Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Stacking the Deck

78 views
Skip to first unread message

Sean Pitman

unread,
Jan 4, 2004, 11:25:02 AM1/4/04
to
lmuc...@yahoo.com (RobinGoodfellow) wrote in message news:<81fa9bf3.04010...@posting.google.com>...
> seanpi...@naturalselection.0catch.com (Sean Pitman) wrote in message news:<80d0c26f.03123...@posting.google.com>...
>
> Good gravy! That was so wrong, it feels wrong to even use the word
> "wrong" to describe it. All I can recommend is that you run, don't
> walk, to your nearest college or university, and sign up as quickly as
> you can for a few math and/or statistics courses: I especially
> recommend courses in probability theory and stochastic modelling.
> With all due respect, Sean, I am beginning to see why the biologists
> and biochemists in this group are so frustrated with you: my
> background in those fields is fairly weak - enough to find your
> arguments unconvincing but not necessarily ridiculous - but if you are
> as weak with biochemistry as you are with statistical and
> computational problems, then I can see why knowledgeable people in
> those areas would cringe at your posts.

With all due respect, what is your area of professional training? I
mean, after reading your post I dare say that you are not only weak in
biology, but statistics as well. Certainly your numbers and
calculations are correct, but the logic behind your assumptions is
extraordinarily fanciful. You sure wouldn't get away with such
assumptions in any sort of peer reviewed medical journal or other
statistically based science journal - that's for sure. Of course, you
may have good success as a novelist . . .

> I'll try to address some of the mistakes you've made below, though I
> doubt that I can do much to dispel your misconceptions. Much of my
> reply will not even concern evolution in a real sense, since I wish to
> highlight and address the mathematical errors that you are making.

What you ended up doing is highlighting your misunderstanding of
probability as it applies to this situation as well as your amazing
faith in an extraordinary stacking of the deck which allows evolution
to work as you envision it working. Certainly, if evolution is true
then you must be correct in your views. However, if you are correct
in your views as stated then it would not be evolution via mindless
processes alone, but evolution via a brilliant intelligently designed
stacking of the deck.

> > RobinGoodfellow <lmuc...@yahoo.com> wrote in message news:<bsd7ue$r1c$1...@news01.cit.cornell.edu>...
>
> > > It is even worse than that. Even random walks starting at random points
> > > in N-dimensional space can, in theory, be used to sample the states
> > > with a desired property X (such as Sean's "beneficial sequences"), even
> > > if the number of such states is exponentially small compared to the
> > > total state space size.
> >
> > This depends upon just how exponentially small the number of
> > beneficial states is relative to the state space.
>
> No, it does not. If you take away anything from this discussion, it
> has to be this: the relative number of beneficial states has virtually
> no bearing on the amount of time a local search algorithm will need to
> find such a state.

LOL - You really don't have a clue how insane this statement is?

> The things that *would* matter are the
> distribution of beneficial states through the state space, the types
> of steps the local search is allowed to take (and the probabilities
> associated with each step), and the starting point.

The distribution of states has very little if anything to do with how
much time it takes to find one of them on average. The starting point
certainly is important to initial success, but it also has very little
if anything to do with the average time needed to find more and more
beneficial functions within that same level of complexity. For
example, if all the beneficial states were clustered together in one
or two areas, the average starting point, if anything, would be
farther way than if these states were distributed more evenly
throughout the sequence space. So, this leaves the only really
relevant factor - the types of steps and the number of steps per unit
of time. That is the only really important factor in searching out
the state space - on average.

> For an extreme
> example, consider a space of strings consisting of length 1000, where
> each position can be occupied by one of 10 possible characters.

Ok. This would give you a state space of 10 to the power of 1000 or
1e1000. That is an absolutely enormous number.

> Suppose there are only two beneficial strings: ABC........, and
> BBC........ (where the dots correspond to the same characters). The
> allowed transitions between states are point mutations, that are
> equally probable for each position and each character from the
> alphabet. Suppose, furthermore, that we start at the beneficial state
> ABC. Then, the probability of a transition from ABC... to BBC... in a
> single mutation 1/(10*1000) = 1/10000 (assuming self-loops - i.e.
> mutations that do not alter the string, are allowed).

You are good so far. But, you must ask yourself this question: What
are the odds that out of a sequence space of 1e1000 the only two
beneficial sequences with uniquely different functions will have a gap
between them of only 1 in 10,000? The time required to cross this
tiny gap would require a random walk of only 10,000 steps on average.
For a decent sized population, this could be done in just one
generation.

Don't you see the problem with this little scenario of yours?
Certainly this is a common mistake made by evolutionists, but it is
none-the less a fallacy of logic. What you have done is assume that
the density of beneficial states is unimportant to the problem of
evolution since it is possible to have the beneficial states clustered
around your starting point. But such a close proximity of beneficial
states is highly unlikely. On average, the beneficial states will be
more widely distributed throughout the sequence space.

For example, say that there are 10 beneficial sequences in this
sequence space of 1e1000. Now say one of these 10 beneficial
sequences just happens to be one change away from your starting point
and so the gap is only a random walk of 10,000 steps as you calculated
above. However, on average, how long will it take to find any one of
the other 9 beneficial states? That is the real question. You rest
your faith in evolution on this inane notion that all of these states
will be clustered around your starting point. If they were, that
certainly would be a fabulous stroke of luck - like it was *designed*
that way. But, in real life, outside of intelligent design, such
strokes of luck are so remote as to be impossible for all practical
purposes. On average we would expect that the other nine sequences
would be separated from each other and our starting point by around
1e999 random walk steps/mutations (i.e., on average it is reasonable
to expect there to be around 999 differences between each of the 10
beneficial sequences). So, even if a starting sequence did happen to
be so extraordinarily lucky to be just one positional change away from
one of the "winning" sequences, the odds are that this luck will not
hold up as well in the evolution of any of the other 9 "winning"
sequences this side of a practical eternity of time.

Real time experiments support this position rather nicely. For
example, a recent and very interesting paper was published by Lenski
et. al., entitled, "The Evolutionary Origin of Complex Features" in
the 2003 May issue of Nature. In this particular experiment the
researchers studied 50 different populations, or genomes, of 3,600
individuals. Each individual began with 50 lines of code and no
ability to perform "logic operations". Those that evolved the ability
to perform logic operations were rewarded, and the rewards were larger
for operations that were "more complex". After only15,873 generations,
23 of the genomes yielded descendants capable of carrying out the most
complex logic operation: taking two inputs and determining if they are
equivalent (the "EQU" function).

In principle, 16 mutations (recombinations) coupled with the three
instructions that were present in the original digital ancestor could
have combined to produce an organism that was able to perform the
complex equivalence operation. According to the researcher themselves,
"Given the ancestral genome of length 50 and 26 possible instructions
at each site, there are ~5.6 x 10e70 genotypes [sequence space]; and
even this number underestimates the genotypic space because length
evolves."

Of course this sequence space was overcome in smaller steps. The
researchers arbitrarily defined 6 other sequences as beneficial (NAND,
AND, OR, NOR, XOR, and NOT functions). The average gap between these
pre-defined steppingstone sequences was 2.5 steps, translating into an
average search space between beneficial sequences of only 3,400 random
walk steps. Of course, with a population of 3,600 individuals in a
population, a random walk of 3,400 will be covered in short order by
at least one member of that population. And, this is exactly what
happened. The average number of mutations required to cross the
16-step gap was only 103 mutations per population.

Now that is lightening fast evolution. Certainly if real life
evolution were actually based on this sort of setup then evolution of
novel functions at all levels of complexity would be a piece of cake.
Of course, this is where most descriptions of this most interesting
experiment stop. But, what the researchers did next is the most
important part of this experiment.

Interestingly enough, Lenski and the other scientists went on to set
up different environments to see which environments would support the
evolution of all the potentially beneficial functions - to include the
most complex EQU function. Consider the following description about
what happened when various intermediate steps were not arbitrarily
defined by the scientists as "beneficial".

"At the other extreme, 50 populations evolved in an environment where
only EQU was rewarded, and no simpler function yielded energy. We
expected that EQU would evolve much less often because selection would
not preserve the simpler functions that provide foundations to build
more complex features. Indeed, none of these populations evolved EQU,
a highly significant difference from the fraction that did so in the
reward-all environment (P = 4.3 x 10e-9, Fisher's exact test).
However, these populations tested more genotypes, on average, than did
those in the reward-all environment (2.15 x 10e7 versus 1.22 x 10e7;
P<0.0001, Mann-Witney test), because they tended to have smaller
genomes, faster generations, and thus turn over more quickly. However,
all populations explored only a tiny fraction of the total genotypic
space. Given the ancestral genome of length 50 and 26 possible
instructions at each site, there are ~5.6 x 10e70 genotypes; and even
this number underestimates the genotypic space because length
evolves."

Isn't that just fascinating? When the intermediate stepping stone
functions were removed, the neutral gap that was created successfully
blocked the evolution of the EQU function, which happened *not* to be
right next door to their starting point. Of course, this is only to
be expected based on statistical averages that go strongly against the
notion that very many possible starting points would just happen to be
very close to an EQU functional sequence in such a vast sequence
space.

Now, isn't this consistent with my predictions? This experiment was
successful because the intelligent designers were capable to defining
what sequences were "beneficial" for their evolving "organisms." If
enough sequences are defined as beneficial and they are placed in just
the right way, with the right number of spaces between them, then
certainly such a high ratio will result in rapid evolution - as we saw
here. However, when neutral non-defined gaps are present, they are a
real problem for evolution. In this case, a gap of just 16 neutral
mutations effectively blocked the evolution of the EQU function.

http://naturalselection.0catch.com/Files/computerevolution.html

> Thus, a random
> walk that restarts each time after the first step (or alternatively, a
> random walk performed by a large population of sequences, each
> starting at state ABC...) is expected to explore, on average, 10000
> states before finding the next beneficial sequence.

Yes, but you are failing to consider the likelihood that your "winning
sequence" will in fact be within these 10,000 steps on average.

> Now, below, we
> will apply your model to the same problem.

Oh, I can hardly wait!

> > It also depends
> > upon how fast this space is searched through. For example, if the
> > ratio of beneficial states to non-beneficial states is as high as say,
> > 1 in a 1e12, and if 1e9 states are searched each second, how long with
> > it take, on average, to find a new beneficial state?
>
> OK. Let's take my example, instead, and apply your calculations.
> There are only 2 beneficial sequences, out of the state space of
> 1e1000 sequences.

Ok, I'm glad that you at least realize the size of the state space.

> Since the ratio of beneficial sequences to
> non-beneficial ones is (2/10^1000), if your "statistics" are correct,
> then I should be exploring 10^1000/2 states, on average, before
> finding the next beneficial state. That is a huge, huge, huge number.
> So why does my very simple random walk explore only 10,000 states,
> when the ratio of beneficial sequences is so small?

Yes, that is the real question and the answer is very simple - You
either got unbelievably lucky in the positioning of your start point
or your "beneficial" sequences were clustered by intelligent design.

> The answer is simple - the ratio of beneficial states does NOT matter!

Yes it does. You are ignoring the highly unlikely nature of your
scenario. Tell me, how often do you suppose your start point would
just happen to be so close to the only other beneficial sequence in
such a huge sequence space? Hmmmm? I find it just extraordinary that
you would even suggest such a thing as "likely" with all sincerity of
belief. The ratio of beneficial to non-beneficial in your
hypothetical scenario is absolutely miniscule and yet you still have
this amazing faith that the starting point will most likely be close
to the only other "winning" sequence in an absolutely enormous
sequence space?! Your logic here is truly mysterious and your faith
is most impressive. I'm sorry, but I just can't get into that boat
with you. You are simply beyond me.

> All that matters is their distribution, and how well a particular
> random walk is suited to explore this distribution.

Again, you must consider the odds that your "distribution" will be so
fortuitous as you seem to believe it will be. In fact, it has to be
this fortuitous in order to work. It basically has to be a set up for
success. The deck must be stacked in an extraordinary way in your
favor in order for your position to be tenable. If such a stacked
deck happened at your table in Las Vegas you would be asked to leave
the casino in short order or be arrested for "cheating" by intelligent
design since such deck stacking only happens via intelligent design.
Mindless processes cannot stack the deck like this. It is
statistically impossible - for all practical purposes.

> (Again, it is a
> gross, meaningless over-simplification to model evolution as a random
> walk over a frozen N-dimensional sequence space, but my point is that
> your calculations are wrong even for that relatively simple model.)

Come now Robin - who is trying to stack the deck artificially in their
own favor here? My calculations are not based on the assumption of a
stacked deck like your calculations are, but upon a more likely
distribution of beneficial sequences in sequence space. The fact of
the matter is that sequence space does indeed contain vastly more
absolutely non-beneficial sequences than it does those that are even
remotely beneficial. In fact, there is an entire theory called the
"Neutral Theory of Evolution". Of all mutations that occur in every
generation in say, humans (around 200 to 300 per generation), the
large majority of them are completely "neutral" and those few that are
functional are almost always detrimental. This ratio of beneficial to
non-beneficial is truly small and gets exponentially smaller with each
step up the ladder of specified functional complexity. Truly,
evolution gets into very deep weeds very quickly beyond the lowest
levels of functional/informational complexity.

> > It will take
> > just over 1,000 seconds - a bit less than 20 minutes on average. But,
> > what happens if at higher levels of functional complexity the density
> > of beneficial functions decreases exponentially with each step up the
> > ladder? The rate of search stays the same, but the junk sequences
> > increase exponentially and so the time required to find the rarer and
> > rarer beneficial states also increases exponentially.
>
> The above is only true if you use the following search algorithm:
>
> 1. Generate a completely random N-character sequence
> 2. If the sequence is beneficial, say "OK";
> Otherwise, go to step 1.

Actually the above is also true if you start with a likely starting
point. A likely starting point will be an average distance away from
the next closest beneficial sequence. A random mutation to a sequence
that does not find the new beneficial sequence will not be selectable
as advantageous and a random walk will begin.

> For an alphabet of size S, where only k characters are "beneficial"
> for each position, the above search algorithm will indeed need to explore
> exponentially many states in N (on average, (S/k)^N), before finding a
> beneficial state. But, this analysis applies only to the above search
> algorithm - an exteremely naive approach that resembles nothing that
> is going on in nature.

Oh really? How do you propose that nature gets around this problem?
How does nature stack the deck so that its starting point is so close
to all the beneficial sequences that otherwise have such a low density
in sequence space?

> The above algorithm isn't even a random walk
> per se, since random walks make local modifications to the current
> state, rather than generate entire states anew.

The random walk I am talking about does indeed make local
modifications to a current sequence. However, if you want to get from
the type of function produced by one state to a new type of function
produced by a different state/sequence, you will need to eventually
leave your first state and move onto the next across whatever neutral
gap there might be in the way. If a new function requires a sequence
that does not happen to be as fortuitously close to your starting
sequence as you like to imagine, then you might be in just a bit of a
pickle. Please though, do explain to me how it is so easy to get from
your current state, one random walk step at a time, to a new state
with a new type of function when the density of beneficial sequences
of the new type of function are extraordinarily infinitesimal?

> A random walk
> starting at a given beneficial sequence, and allowing certain
> transitions from one sequence to another, would require a completely
> different type of analysis. In the analyses of most such search
> algorithms, the "ratio" of beneficial sequences would be irrelevant -
> it is their *distribution* that would determine how well such an
> algorithm would perform.

The most likely distribution of beneficial sequences is determined by
their density/ratio. You cannot simply assume that the deck will be
so fantastically stacked in the favor of your neat little evolutionary
scenario. I mean really, if the deck was stacked like this with lots
of beneficial sequences neatly clustered around your starting point,
evolution would happen very quickly. Of course, there have been those
who propose the "Baby Bear Hypothesis". That is, the clustering is
"just right" so that the theory of evolution works. That is the best
you can hope for. Against all odds the deck was stacked just right so
that we can still believe in evolution. Well, if this were the case
then it would still be evolution by design. Mindless processes just
can't stack the deck like you are proposing.

> My example above demonstrates a problem
> where the ratio of beneficial states is exteremely tiny, yet the
> search finds a new beneficial state relatively quickly.

Yes - because you stacked the deck in your favor via deliberate
design. You did not even try to explain the likelihood of this
scenario in real life. How do you propose that this is even a remote
reflection of what mindless processes are capable of? I'm talking
average probabilities here while you are talking about extraordinarily
unlikely scenarios that are basically impossible outside of deliberate
design.

> I could also
> very easily construct an example where the ratio is nearly one, yet a
> random walk starting at a given beneficial sequence would stall with a
> very high probability.

Oh really? You can construct a scenario where all sequences are
beneficial and yet evolution cannot evolve a new one? Come on now . .
. now you're just being silly. But I certainly would like to see you
try and set up such a scenario. I think it would be most
entertaining.

> In other words, Sean, your calculations are
> irrelevant for the kind of problem you are trying to analyze.

Only if you want to bury your head in the sand and force yourself to
believe in the fairytale scenarios that you are trying to float.

> If you
> wish to model evolution as a random walk of point mutations on a
> frozen N-dimensional sequence space, you will need to apply a totally
> different statististical analysis: one that takes into account the
> distributions of known "beneficial" sequences in sequence space. And
> then I'll tell you why that model too is so wrong as to be totally
> irrelevant.

And if you wish to model evolution as a walk between tight clusters of
beneficial sequences in an otherwise extraordinarily low density
sequence space, then I have some oceanfront property in Arizona to
sell you at a great price.

Until then, this is all I have time for today.

> Cheers,
> RobinGoodfellow.

Sean
www.naturalselection.0catch.com

"Rev Dr" Lenny Flank

unread,
Jan 4, 2004, 1:48:24 PM1/4/04
to
Sean Pitman wrote:


>
> Until then, this is all I have time for today.


Hey doc, when will you have time to tell us what the scientific theory
of intelligent design is --- what does the desigher do, specifically,
what mechanisms does it use to do it, where can we see these mechanisms
in operation today. And what idnicates there is only one desinger and
not, say, ten or fifty of them all working together.

After that, can you find the time to explain to me how ID "theory" is
any less "materialist" or "naturalist" or "atheist" than is evolutionary
biology, since ID "theory" not only does NOT hypothesize the existence
of any supernatural entities or actions, but specifically states that
the "intelligent designer" might be nothing but a space alien.

And after THAT, could you find the time to tell us how you apply
anything other than "naturalism" or "materialism" to your medical
practice? What non-naturalistic cures do you recommend for your
patients, doctor.

I do understand that you wont' answer, doc. That's OK. The questions
make their point -- with you or without you.

===============================================
Lenny Flank
"There are no loose threads in the web of life"

Creation "Science" Debunked:
http://www.geocities.com/lflank

DebunkCreation Email list:
http://www.groups.yahoo.com/group/DebunkCreation

-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----

RobinGoodfellow

unread,
Jan 4, 2004, 10:53:13 PM1/4/04
to
I've already responded to this same post in a different thread. See:

http://groups.google.com/groups?dq=&hl=en&lr=&ie=UTF-8&threadm=3FF89BDA.EB18D013%40indiana.edu&prev=/groups%3Fdq%3D%26num%3D25%26hl%3Den%26lr%3D%26ie%3DUTF-8%26group%3Dtalk.origins%26start%3D50
or
http://makeashorterlink.com/?C309615F6

Incidentally, I'll be leaving for a much-needed vacation in a couple
of days, and expect that other commitments will force me to return to
lurkdom for a while afterwards. So I apologize in advance for leaving
these two threads hanging, though I look forward to reading your
replies.

Cheers,
Robin.


seanpi...@naturalselection.0catch.com (Sean Pitman) wrote in message news:<80d0c26f.04010...@posting.google.com>...

Sean Pitman

unread,
Jan 14, 2004, 6:02:24 AM1/14/04
to
RobinGoodfellow <lmuc...@yahoo.com> wrote in message news:<bt8i6p$r9h$1...@news01.cit.cornell.edu>...

> > Sean Pitman wrote:
> >
> > With all due respect, what is your area of professional training? I
> > mean, after reading your post I dare say that you are not only weak in
> > biology, but statistics as well. Certainly your numbers and
> > calculations are correct, but the logic behind your assumptions is
> > extraordinarily fanciful. You sure wouldn't get away with such
> > assumptions in any sort of peer reviewed medical journal or other
> > statistically based science journal - that's for sure. Of course, you
> > may have good success as a novelist . . .
>
> Tsk, tsk... I thank you for the career advice. I'll keep it in mind,
> should my current stint in computer science fall through. I wouldn't go
> so far as to say that Monte-Carlo methods are my specialty, but I will
> say that my own research and the research of half my colleagues would be
> non-existent if they worked the way you think they do.

Hmmmm, so what has your research shown? I've seen nothing from the
computer science front that shows how anything new, such as a new
software program, beyond the lowest levels of functional complexity
can be produced by computers without the input of an intelligent mind.
Your outlandish claims for the result of research done so far, such
as the Lenski experiments, are just over the top. They don't
demonstrate anything even close to what you claim they demonstrate
(See Below).

> >>I'll try to address some of the mistakes you've made below, though I
> >>doubt that I can do much to dispel your misconceptions. Much of my
> >>reply will not even concern evolution in a real sense, since I wish to
> >>highlight and address the mathematical errors that you are making.
> >
> > What you ended up doing is highlighting your misunderstanding of
> > probability as it applies to this situation as well as your amazing
> > faith in an extraordinary stacking of the deck which allows evolution
> > to work as you envision it working. Certainly, if evolution is true
> > then you must be correct in your views. However, if you are correct
> > in your views as stated then it would not be evolution via mindless
> > processes alone, but evolution via a brilliant intelligently designed
> > stacking of the deck.
>

> Exactly what views did I state, Sean? Other than that your calculations
> are, to put it plainly, irrelevant. Not even wrong - just irrelevant.
>
> Yes, the example I give below incredibly stacks the deck in my favor.
> It ought to. It is what is called a "counter-example". It falsifies
> the hypothesis that your "model" of evolution is correct. Now aren't
> you glad you proposed something falsifiable?

Come again? How does your stacking the deck via the use of
intelligent design, since there is no other logical way to stack the
deck so that your scenario will actually work, disprove my position?
My hypothesis is dependent on the far more likely scenario that the
deck in not stacked as you suggest, but is in fact much more random
than you seem to think it is. Certainly the ONLY way evolution could
work is if the deck was stacked, but then this would be easily
detected as evidence of intelligent design, not the normal
understanding of evolution as a mindless non-directed process.

> > This distribution of states has very little if anything to do with how


> > much time it takes to find one of them on average. The starting point
> > certainly is important to initial success, but it also has very little
> > if anything to do with the average time needed to find more and more
> > beneficial functions within that same level of complexity.
>

> Except in every real example of a working Monte-Carlo procedure, where
> the distribution and starting point have *everything* to do whether such
> a procedure is successful or not.

You mean that the stacking of the deck has everything to do with
whether or not an "evolutionary" scenario will succeed. Certainly
this would be true, but such a stacking of the deck has no resemblance
to reality. You must ask yourself about the likelihood that one will
find such a stacked deck in real life outside of intelligent design .
. .

> > For
> > example, if all the beneficial states were clustered together in one
> > or two areas, the average starting point, if anything, would be
> > farther way than if these states were distributed more evenly
> > throughout the sequence space. So, this leaves the only really
> > relevant factor - the types of steps and the number of steps per unit
> > of time. That is the only really important factor in searching out
> > the state space - on average.
>

> *Sigh*. The problem is that the model *you* are proposing (one I think
> is silly) is of a random on walk on a specific frozen sequence space
> with beneficial sequences as points in that space. It does not deal
> with an "average" distribution, and an "average" starting point, but
> with one very specific distribution of beneficial sequences and one very
> specific starting point.

Consider the scenario where there are 10 ice cream cones on the
continental USA. The goal is for a blind man to find as many as he
can in a million years. It seems that what you are suggesting is that
the blind man should expect that the ice cream cones will all be
clustered together and that this cluster will be with arms reach of
where he happens to start his search. This is simply a ludicrous
notion outside of intelligent design. My hypothesis, on the other
hand, suggests that these 10 ice cream cones will have a more random
distribution with hundreds of miles separating each one, on average.
An average starting point of the blind man may, by a marvelous stroke
of luck, place him right beside one of the 10 cones. However, after
finding this first cone, how long, on average, will it take him to
find any of the other 9 cones? That is the question here. The very
low density of ice cream cones translates into a marked increase in
the average time required to find them. Now, if there were billions
upon billions of ice cream cones all stuffed into this same area, then
one could reasonably expect that they would be separated by a much
closer average distance - say just a couple of feet. With such a high
density, the average time needed for the blind man to find another ice
cream cone would be just a few seconds.

So, whose position is more likely? Your notion that the density of
beneficial sequences in sequence space doesn't matter or my notion
that density does matter? Is your hypothetical situation where a low
density of beneficial states is clustered around a given starting
point really valid outside of intelligent design? If so, name a
non-designed situation where such an unlikely phenomenon has ever been
observed to occur . . .

> You cannot simply assume an "average"
> distribution in the absence of background information: you have to find
> out precisely the kind of distribution you are dealing with. And even
> if you do find that the distribution is "stacked", it does not imply
> that an intelligence was involved.

Oh really? You think that stacking the deck as you have done can
happen mindlessly in less than zillions of years of average time?
Come on now! What planet are you from?

> The stacking could occur due to the
> constraints imposed by the very definition of the problem: in the case
> of evolutions, by the physical constraints governing the interactions
> between the molecules involved in biological systems.

Oh, so the physical laws of atoms and molecules force them to
self-assemble themselves in functionally complex systems? Now you are
really reaching. Tell me why the physical constraints of these
molecular machines force all beneficial possibilities to be so close
together? This is simply the most ludicrous notion that I have heard
in a very long time. You would really do well in Vegas with that one!
Try telling them, when they come to arrest you for cheating, that the
deck was stacked because of the physical constraints of the playing
cards.

> In fact, why
> would you expect that the regular and highly predictable physical laws
> governing biochemical reactions would produce a random, "average"
> distribution of "beneficial sequences"?

Because, I don't know of any requirement for them to be clustered
outside of deliberate design - do you? I can see nothing special
about the building blocks that make up living things that would cause
the potentially beneficial systems found in living things to have to
be clustered (just like there is nothing inherent in playing cards
that would cause them to stack themselves in any particular order).
However, if you know of a reason why the physical nature of the
building blocks of life would force them to cluster together despite
having a low density in sequence space, please, do share it with me.
Certainly none of your computer examples have been able to demonstrate
such a necessity. Why then would you expect such a forced clustering
in the potentially beneficial states of living things?

> >>For an extreme
> >>example, consider a space of strings consisting of length 1000, where
> >>each position can be occupied by one of 10 possible characters.
>

> Note, I wrote, "extereme example". My point was *not* invent a
> distribution which makes it likely for evolutiuon to occur (this example
> has about as much to do with evolution as ballet does with quantum
> mechanics), but to show how inadequate your methods are.

Actually, this situation has a lot to do with evolution and is the
real reason why evolution is such a ludicrous idea. What your
illustration shows is that only if the deck is stacked in a most
unlikely way will evolution have the remotest possibility of working.
That is what I am trying to show and you demonstrated this very
nicely. Unwittingly it is you who effectively show just how
inadequate evolutionary methods are at making much of anything outside
of an intelligently designed stacking of the deck.



> >>Suppose there are only two beneficial strings: ABC........, and
> >>BBC........ (where the dots correspond to the same characters). The
> >>allowed transitions between states are point mutations, that are
> >>equally probable for each position and each character from the
> >>alphabet. Suppose, furthermore, that we start at the beneficial state
> >>ABC. Then, the probability of a transition from ABC... to BBC... in a
> >>single mutation 1/(10*1000) = 1/10000 (assuming self-loops - i.e.
> >>mutations that do not alter the string, are allowed).
> >
> >
> > You are good so far. But, you must ask yourself this question: What
> > are the odds that out of a sequence space of 1e1000 the only two
> > beneficial sequences with uniquely different functions will have a gap
> > between them of only 1 in 10,000?
>

> Mind-numbingly low. 1000*.9*.1^999, to be precise. But that is not the
> point.

Actually, this is precisely the point. What you are basically saying
is that if there were only one ice cream cone in the entire universe
that it could be easily found if the starting point of the blind man's
search just so happened to be an arms reach away from the cone. That
is what you are saying is it not?



> > Don't you see the problem with this little scenario of yours?
> > Certainly this is a common mistake made by evolutionists, but it is
> > none-the less a fallacy of logic. What you have done is assume that
> > the density of beneficial states is unimportant to the problem of
> > evolution since it is possible to have the beneficial states clustered
> > around your starting point. But such a close proximity of beneficial
> > states is highly unlikely. On average, the beneficial states will be
> > more widely distributed throughout the sequence space.
>

> On average, yes.

On average yes?! How can you say this and yet disagree with my
conclusions?

> But didn't you just say above that the distribution
> of the sequences is irrelevant? That all that matters is "ratio" of
> beneficial sequences?

It is only by determining the ration of beneficial sequences that you
can obtain a reasonable idea about the likely distribution of these
sequences around any particular starting point. You hold a huge
fallacy of logic that by some magical means the distribution could be
just right even though the density is truly miniscule (like the
finding one atom in zillions of universes the size of ours).

> (Incidentally, "ratio" and "density" are not
> identical. The distribution I showed you has a relatively high density
> of beneficial sequences, despite a low ratio.)

You are talking local "density", which, in your scenario, also has a
locally high "ratio". I, on the other hand, was talking about the
total ratio and density of the whole potential space taken as a whole.
Really, you are very much mistaken to suggest that the ratio and
density of a state in question per the same unit of state space are
not equivalent.

> > For example, say that there are 10 beneficial sequences in this
> > sequence space of 1e1000. Now say one of these 10 beneficial
> > sequences just happens to be one change away from your starting point
> > and so the gap is only a random walk of 10,000 steps as you calculated
> > above. However, on average, how long will it take to find any one of
> > the other 9 beneficial states? That is the real question. You rest
> > your faith in evolution on this inane notion that all of these states
> > will be clustered around your starting point. If they were, that
> > certainly would be a fabulous stroke of luck - like it was *designed*
> > that way. But, in real life, outside of intelligent design, such
> > strokes of luck are so remote as to be impossible for all practical
> > purposes. On average we would expect that the other nine sequences
> > would be separated from each other and our starting point by around
> > 1e999 random walk steps/mutations (i.e., on average it is reasonable
> > to expect there to be around 999 differences between each of the 10
> > beneficial sequences). So, even if a starting sequence did happen to
> > be so extraordinarily lucky to be just one positional change away from
> > one of the "winning" sequences, the odds are that this luck will not
> > hold up as well in the evolution of any of the other 9 "winning"
> > sequences this side of a practical eternity of time.
>

> Unless, of course, it follows from the properties of the problem that
> the other 9 benefecial sequences must be close to the starting sequence.

And I am sure you have some way to explain why these 9 other
beneficial sequences would have to be close together outside of
deliberate design? What "properties" of the problem would force such
a low density of novel beneficial states to be so clustered? I see
absolutely no reason to suggest such a necessity. Certainly such a
necessity much be true if evolution is true, but if no reasonable
naturalistic explanation can be given, why should I simply assume such
a necessity? Upon what basis do you make this claim?

> > Real time experiments support this position rather nicely. For
> > example, a recent and very interesting paper was published by Lenski
> > et. al., entitled, "The Evolutionary Origin of Complex Features" in
> > the 2003 May issue of Nature. In this particular experiment the
> > researchers studied 50 different populations, or genomes, of 3,600
> > individuals. Each individual began with 50 lines of code and no
> > ability to perform "logic operations". Those that evolved the ability
> > to perform logic operations were rewarded, and the rewards were larger
> > for operations that were "more complex". After only15,873 generations,
> > 23 of the genomes yielded descendants capable of carrying out the most
> > complex logic operation: taking two inputs and determining if they are
> > equivalent (the "EQU" function).
>

> I've already covered how you've completely misinterpreted Lenski's
> research in the other post. But let's run with this for a bit:

Lets . . . Oh, and if you would give a link to where you "covered" my
"misinterpretation", that would be appreciated.

> > In principle, 16 mutations (recombinations) coupled with the three
> > instructions that were present in the original digital ancestor could
> > have combined to produce an organism that was able to perform the
> > complex equivalence operation. According to the researcher themselves,
> > "Given the ancestral genome of length 50 and 26 possible instructions
> > at each site, there are ~5.6 x 10e70 genotypes [sequence space]; and
> > even this number underestimates the genotypic space because length
> > evolves."
> >
> > Of course this sequence space was overcome in smaller steps. The
> > researchers arbitrarily defined 6 other sequences as beneficial (NAND,
> > AND, OR, NOR, XOR, and NOT functions).
>

> As a minor quibble, I believe they actually started with NAND (you need
> it for all the other functions). But I could be wrong - I've read that
> paper months ago.

You are correct. The fact is though that the NAND starting point was
defined as beneficial and it was not made up of random sequences of
computer code. It was all set up very specifically so that certain
recombinations of code (point mutations were not primarily used,
though they did happen on occasion during recombination events), would
yield certain types of other pre-determined coded functions.

> > those in the reward-all environment (2.15 x 1e7 versus 1.22 x 1e7;


> > P<0.0001, Mann-Witney test), because they tended to have smaller
> > genomes, faster generations, and thus turn over more quickly. However,
> > all populations explored only a tiny fraction of the total genotypic
> > space. Given the ancestral genome of length 50 and 26 possible

> > instructions at each site, there are ~5.6 x 1e70 genotypes; and even


> > this number underestimates the genotypic space because length
> > evolves."
>

> And after years of painstaking research, Sean finally invents the wheel.
> Yes, evolution does not pop complex systems out of thin air, but
> constructs through integration and co-optation of simpler functional
> components. Move along, folks, nothing to see here!

What this shows is that if the "simpler" components aren't defined as
"beneficial" then a system of somewhat higher complexity will not
evolve at all - period - even given zillions of years of time. Truly,
this means that there really isn't anything to see here. Nothing
evolves without the deck being stacked by intelligent design. That is
all this Lenski experiment showed.

> > Isn't that just fascinating? When the intermediate stepping stone
> > functions were removed, the neutral gap that was created successfully
> > blocked the evolution of the EQU function, which happened *not* to be
> > right next door to their starting point. Of course, this is only to
> > be expected based on statistical averages that go strongly against the
> > notion that very many possible starting points would just happen to be
> > very close to an EQU functional sequence in such a vast sequence
> > space.
>

> Here's a question for you. There were only 5 beneficial functions in
> that big old sequence space of yours.

Actually, including the starting and ending points, there were 7
defined beneficial sequences in this sequence space (NAND, AND, OR,
NOR, XOR, NOT, and EQU functions).

> They are all very standard
> Boolean functions: in no way were they specifically designed by Lenski
> et. al. to ease the way to into evolving the EQ functions.

Actually, they very much were designed by Lenski et. al. to ease the
way along the path to the EQU sequence. The original code was set up
with very specific lines of code that could, when certain
recombinations occurred, give rise to each of these logic functions.
The lines of code were not random lines of code and they were not all
needed to be as they were for the original NAND function to operate.
In fact the researchers knew the approximate rate of evolution that
would be expected ahead of time based on their programming of the
coded sequences, the rate of recombination of these sequences, the
size of the sequence space and the distance between each step along
the pathway. It really was a very nice setup for success. Read the
paper again and you will see that this is true.

> How come
> they were all sufficiently close in sequence space to one another, when
> according to you such a thing is so highly improbable?

Because they were designed to be close together deliberately. The
deck was stacked on purpose. I mean really, you can't be suggesting
that these 7 beneficial states just happened to be clustered together
in a state space of 1e70 by the mindless restriction of the program do
you? The program was set up with the restrictions stacked in a
particular way so that only these 7 states could evolve and that each
subsequence state was just a couple of steps away from the current
state. No other function was set up to evolve, so no other novel
function evolved. These lines of code did not get together and make a
calculator program or a photo-editing program, or even a simple
program to open the CD player. That should tell you something . . .
This Lenski experiment was *designed* to succeed like it did. Without
such input of intelligent deck stacking, it never would have worked
like it did.

> > Now, isn't this consistent with my predictions? This experiment was
> > successful because the intelligent designers were capable to defining
> > what sequences were "beneficial" for their evolving "organisms." If
> > enough sequences are defined as beneficial and they are placed in just
> > the right way, with the right number of spaces between them, then
> > certainly such a high ratio will result in rapid evolution - as we saw
> > here. However, when neutral non-defined gaps are present, they are a
> > real problem for evolution. In this case, a gap of just 16 neutral
> > mutations effectively blocked the evolution of the EQU function.
>

> You are not even close. Lenski et. al. didn't define which *sequences*
> were "beneficial".

Yes, they did exactly that. Read the paper again. They arbitrarily
wrote the code in a meaningful way for the starting lines as well as
arbitrarily defined which recombinations would be "beneficial". They
say it in exactly that way. They absolutely say that they defined
what was and what was not "beneficial".

> They didn't even design functions to serve
> specifically as stepping stones in the evolutionary pathways of EQ.

Yes they did in that they wrote the original code so that it would be
possible to form such pre-defined "beneficial" codes in a series of
recombinations of lines of code.

> What they have done is to name some functions of intermediate complexity
> that might be beneficial to the organism.

You obviously either haven't read the original paper or you don't
understand what it said. The researchers openly admit to arbitrarily
defining the "intermediate" states as beneficial. This fact is only
proven because they went on to remove the "beneficial" definition from
these intermediate states. Without this arbitrary assignment of
beneficial to the intermediate states, the EQU state did not evolve.
Go back an read the paper again. It was the researchers who defined
the states. The states themselves obviously didn't have inherent
benefits in the "world" that they were evolving in outside of the
researcher's definitions for them.

> They certainly did not tell
> their program how to reach these functions, or what the systems
> performing these functions might look like, but simply indicated that
> there are functions at varying levels of complexity that might be useful
> to an organism in its environment.

Wrong again. They did in fact tell their program exactly which
states, specifically, to reward and how to reward them if present.
They told the program exactly what they would look like ahead of time
so that they would be recognized and treated as beneficial when they
arrived on the scene.

You really don't seem like you have a clue how this experiment was
done. I really don't understand how you can make such statements as
this if you had actually read the paper.

> Thus, they have demonstrated exactly
> what they set out to: that in evolution, complex functional features are
> acquired through co-optation and modification of simpler ones.

They did nothing of the sort. All they did was show that stacking the
deck by intelligent design really does work. The problem is that
evolution is supposed to work to create incredible diversity and
informational complexity without any intelligent intervention having
ever been required. So, you evolutionists are back to ground zero.
There simply is no evolution, outside of intelligent design, beyond


the lowest levels of functional/informational complexity.

<snip>


> >>(Again, it is a
> >>gross, meaningless over-simplification to model evolution as a random
> >>walk over a frozen N-dimensional sequence space, but my point is that
> >>your calculations are wrong even for that relatively simple model.)
> >
> > Come now Robin - who is trying to stack the deck artificially in their
> > own favor here? My calculations are not based on the assumption of a
> > stacked deck like your calculations are, but upon a more likely
> > distribution of beneficial sequences in sequence space. The fact of
> > the matter is that sequence space does indeed contain vastly more
> > absolutely non-beneficial sequences than it does those that are even
> > remotely beneficial.
>

> Yes, but your caclulations are based on the equally unfounded assumption
> that the deck is not stacked in any way, shape, or form. (That is, if
> the sequences were really distributed evenly in your frozen sequence
> space, then your probability calculation would still be off, but not by
> too much.)

Not by too much? Hmmmmm . . . So, you are saying that if the
sequence space where set up even close to the way in which I am
suggesting then my calculations would be pretty much correct? So,
unless the sequence space looks like you envision it looking, all nice
and neatly clustered around your pre-arranged starting point, then I
am basically right? So, either the deck is stacked pretty much like
you suggest or the deck is more randomly distributed like I suggest.
If it is stacked, then you are correct and evolution is saved. If the
deck is more randomly distributed like I suggest, then evolution is
false and should be discarded as untenable - correct?

Now where did I miss it? You said at the beginning that my
calculations were completely off base given my own position and that
you were going to correct my math. You said that I needed special
training in statistics. Now, how can my calculations be pretty much
on target given my hypothesis and yet I not know anything about
statistics?

> What makes you think that the laws of physics do not stack
> the deck sufficiently to make evolution possible?

More importantly, what makes you think that they do? I've never seen
a mindless process stack the deck like this, have you? Where are your
examples of mindless processes stacking the deck in such as way as you
suggest outside of aid of intelligent design?

> You may feel that
> they can't: but in the meantime, you should be striving to find out what
> the actual distribution is, rather than assuming it is unstacked. (Not
> that this would make your model relevant, but it'll be a small step in
> the right direction.)

Actually, an unstacked deck would make my model very relevant indeed.
You admit as much yourself when you say that my calculations are
pretty much correct give that the hypothesis of an unstacked deck is
true. Now, the ball is in your court. It is so extremely
counterintuitive to me that the deck would be unstacked that such an
assertion demands equivalent evidence. Where do you see such deck
stacking outside of intelligent design? That is the real question
here.

> > In fact, there is an entire theory called the
> > "Neutral Theory of Evolution". Of all mutations that occur in every
> > generation in say, humans (around 200 to 300 per generation), the
> > large majority of them are completely "neutral" and those few that are
> > functional are almost always detrimental. This ratio of beneficial to
> > non-beneficial is truly small and gets exponentially smaller with each
> > step up the ladder of specified functional complexity. Truly,
> > evolution gets into very deep weeds very quickly beyond the lowest
> > levels of functional/informational complexity.
>

> The fact that the vast majority of mutations are neutral does not imply
> that there exists any point where there is no opportunity for a
> beneficial mutation. And where such an opportunity presents itself,
> evolution will eventually find it, given large enough populations and
> sufficient times.

Yes, if by "sufficient time" you mean zillions of years - even for
extremely large populations.

> >>>It will take
> >>>just over 1,000 seconds - a bit less than 20 minutes on average. But,
> >>>what happens if at higher levels of functional complexity the density
> >>>of beneficial functions decreases exponentially with each step up the
> >>>ladder? The rate of search stays the same, but the junk sequences
> >>>increase exponentially and so the time required to find the rarer and
> >>>rarer beneficial states also increases exponentially.
> >>
> >>The above is only true if you use the following search algorithm:
> >>
> >> 1. Generate a completely random N-character sequence
> >> 2. If the sequence is beneficial, say "OK";
> >> Otherwise, go to step 1.
> >
> > Actually the above is also true if you start with a likely starting
> > point. A likely starting point will be an average distance away from
> > the next closest beneficial sequence. A random mutation to a sequence
> > that does not find the new beneficial sequence will not be selectable
> > as advantageous and a random walk will begin.
>

> Actually, your last paragraph will be approximately true only if all
> your "beneficial" points are uniformly spread out through your sequence
> space.

In other words, if they aren't stacked in some extraordinarily
fortuitous fashion?

> Even then, your probability calculation will be off by some
> orders of magnitude, since you will actually need to apply combinatorial
> forumlas to compute these probabilities correctly. But, I suppose,
> it'll be close enough.

My calculations will not be off too far. And, even if they are off by
a few orders of magnitude, it doesn't matter compared to the numbers
involved. As you say, the rough estimates involved here are clearly,
"close enough" to get a very good idea of the problem. My math is not
"way off" as you originally indicated. If anything you have a
conceptual problem with my hypothesis, not my statistics/math. It
basically boils down to this: Either the deck was stacked by a
mindless or a mindful process. You have yet to provide any convincing
evidence that a mindless process can stack a deck, like it would have
to have been stacked for life forms to be as diverse and complex they
are, outside of a lot of help from intelligent design.

<snip>


> >> I could also
> >>very easily construct an example where the ratio is nearly one, yet a
> >>random walk starting at a given beneficial sequence would stall with a
> >>very high probability.
> >
> > Oh really? You can construct a scenario where all sequences are
> > beneficial and yet evolution cannot evolve a new one? Come on now . .
> > . now you're just being silly. But I certainly would like to see you
> > try and set up such a scenario. I think it would be most
> > entertaining.
>

> I didn't say all sequences are beneficial, Sean. That *would* be silly.
> I did say that the ratio *approaches* one, but is not quite that.
> But, here you are:
>
> Same "sequence space" as before, but now a sequence is "beneficial" if
> it is AAAAAAAAAA......AAA (all A's), or it differs from AAAAA...AAA by
> at least 2 amino acids. All other sequences are *harmful* - if the
> random walk ever stumbles onto one, it will die off, and will need to
> return to its starting point. (This means there are exactly 1000*9 +
> (1000*999/2)*81 or about 4.02e6 harmful sequences, and 1e1000-4.02e6 or
> about 1e1000 beneficial sequences: that is, virtually every sequence is
> beneficial.) Again, the allowed transitions are point mutations, and
> the starting point is none other AAAAAAA...AAA. Now, will this random
> walk ever find another beneficial sequence?

Your math here seems to be just a bit off. For example, if out of
1e1000 the number of beneficial sequences was 1e999, the ratio of
beneficial sequences would be 1 in 10. At this ratio, the average
distance to a new beneficial function would not be "two amino acid
changes away", but less than one amino acid change away. The ratio
created by "at least 2 amino acid changes" is less than 1 in 400, not
less than 1 in 10 like you suggest here.

Also, even if all sequences less than 2 amino acid changes were
detrimental (which is very unlikely), an average bacterial colony of
100 billion or so individuals would cross this 2 amino acid gap in
short order since a colony this size would experience a double
mutation in a sequence this size in several members of its population
during the course of just one generation.



> > And if you wish to model evolution as a walk between tight clusters of
> > beneficial sequences in an otherwise extraordinarily low density
> > sequence space, then I have some oceanfront property in Arizona to
> > sell you at a great price.
>

> If I did wish to model evolution this way, then I would gladly buy this
> property off your hands. And then sell it back to you at twice the
> price, because it would still be better than the model you propose.

LOL - Ok, you just keep thinking that way. But, until you have some
evidence to support your wishful thinking mindless stacking of the
deck hypothesis, what is there to make your position attractive or
even remotely logical?

> Cheers,
> RobinGoodfellow.

Sean
www.naturalselection.0catch.com

Chris Merli

unread,
Jan 14, 2004, 10:00:08 AM1/14/04
to

"Sean Pitman" <seanpi...@naturalselection.0catch.com> wrote in message
news:80d0c26f.04011...@posting.google.com...

But there is not one blind man looking there are many and only those close
enough to the cluster of cones in the first place are likely to succeed.

>
> So, whose position is more likely? Your notion that the density of
> beneficial sequences in sequence space doesn't matter or my notion
> that density does matter? Is your hypothetical situation where a low
> density of beneficial states is clustered around a given starting
> point really valid outside of intelligent design? If so, name a
> non-designed situation where such an unlikely phenomenon has ever been
> observed to occur . . .
>
> > You cannot simply assume an "average"
> > distribution in the absence of background information: you have to find
> > out precisely the kind of distribution you are dealing with. And even
> > if you do find that the distribution is "stacked", it does not imply
> > that an intelligence was involved.
>
> Oh really? You think that stacking the deck as you have done can
> happen mindlessly in less than zillions of years of average time?
> Come on now! What planet are you from?

Lets talk clusters. How many point mutations of a protein are in fact still
functional. This tends to create a cluster all of its own. Given this fact
the idea that they are spread evenly accross the landscape is just not true.

howard hershey

unread,
Jan 14, 2004, 2:52:27 PM1/14/04
to

Sean Pitman wrote:

Except that is NOT what evolution does. Evolution starts with an
organism with pre-existing sequences that produce products and interact
with environmental chemicals in ways that are useful to the organism's
reproduction. The situation is more like 10,000 blind men in a varying
topography who blindly follow simple and dumb rules of the game to find
useful things (ice cream at the tops of fitness peaks): Up is good. Down
is bad. Flat is neither good nor bad. Keep walking in all cases. It
would not take too long for these 10,000 blind men to be found in
decidedly non-random places (the high mesas of functional utility where
they are wandering around the flat tops if you haven't guessed). And
the ice cream cones (the useful functions), remember, are not randomly
distributed either. They are specifically at the tops of these mesas as
well. That is what a fitness landscape looks like.

If this topography of utility only changed slowly, at any given time it
would appear utterly amazing to Sean that the blind men will all be
found at these local high points or optimal states (the mesas licking
the ice cream cones on them) rather than being randomly scattered around
the entire surface. They reached these high points (with the ice cream)
by following a simple dumb algorithm.

But you were wondering how something new could arise *after* the blind
men are already wandering around the mesas? The answer is that it
depends. They can't always do so. But remember that these pre-existing
mesas are not random places. They do something specific with local
utility. Let's say that each mesa top has a different basic *flavor* of
ice cream. Say that chocolate is a glycoside hydrolase that binds a
glucose-based glycoside. Now let's say that the environment changes so
that one no longer needs this glucose-based glycoside (the mesa sinks
down to the mean level) but now one needs a galactose-based glycoside
hydrolase. Notice that the difference in need here is something more
like wanting chocolate with almonds than wanting even strawberry, much
less jalapeno or anchovy-flavored ice cream. The blind man on the newly
sunk mesa must keep walking, of course, but he is not thousands of miles
away from the newly risen mesa with chocolate with almonds ice cream on
top. Changing from one glucose-based glycoside hydrolase to one with a
slightly different structure is not the same as going from chocolate to
jalapeno or fish-flavored ice cream. Not even the same as going from
chocolate to coffee. The "island" of chocolate with almonds is *not*
going to be way across the ocean from the "island" of chocolate. It will
be nearby where the blind man is. *And* because chocolate with almonds
is now the need, it will also be on the new local high mesa (relative to
the position of the blind man on the chocolate mesa). The blind man
need only follow the simple rules (Up good. Down bad. Neutral neutral.
Keep walking.) and he has a good chance of reach the 'new' local mesa
top quite often.

And remember that there is not just one blind man on one mesa in this
ocean of possible sequences. There are 10,000 already present on 10,000
different local mesas with even more flavors than the 31 that most ice
cream stores offer. Your math always presuposes that whenever you need
to find, say, vanilla with cherry the one blind man starts in some
random site and walks in a completely random fashion (rather than by the
rules I pointed out) across half the universe of sequence space to reach
your pre-determined goal by pure dumb luck to find the perfect lick. My
presumption is that the successful search is almost always going to
start from the pre-existing mesa with the closest flavor to the new need
(or from a duplicate, which, as a duplicate, is often superfluous and
quickly erodes to ground level in terms of its utility). As mentioned,
these pre-existing mesas are not random pop-ups. They are at the most
useful places in sequence space from which to try to find near-by mesas
with closely-related biologically useful properties because they already
have biologically useful properties.

> It seems that what you are suggesting is that
> the blind man should expect that the ice cream cones will all be
> clustered together and that this cluster will be with arms reach of
> where he happens to start his search. This is simply a ludicrous
> notion outside of intelligent design. My hypothesis, on the other
> hand, suggests that these 10 ice cream cones will have a more random
> distribution with hundreds of miles separating each one, on average.
> An average starting point of the blind man may, by a marvelous stroke
> of luck, place him right beside one of the 10 cones. However, after
> finding this first cone, how long, on average, will it take him to
> find any of the other 9 cones? That is the question here. The very
> low density of ice cream cones translates into a marked increase in
> the average time required to find them. Now, if there were billions
> upon billions of ice cream cones all stuffed into this same area, then
> one could reasonably expect that they would be separated by a much
> closer average distance - say just a couple of feet. With such a high
> density, the average time needed for the blind man to find another ice
> cream cone would be just a few seconds.
>
> So, whose position is more likely?

Your position is not wrong. It is simply irrelevant and unrelated to
reality.

> Your notion that the density of
> beneficial sequences in sequence space doesn't matter or my notion
> that density does matter?

All that matters is whether there is a pre-existing sequence close
enough to one that meets your requirement for being beneficial. And
pre-existing sequences in biological organisms are not random. And
there are more than one such sequence. The only one that matters is the
closest one.

> Is your hypothetical situation where a low
> density of beneficial states is clustered around a given starting
> point really valid outside of intelligent design? If so, name a
> non-designed situation where such an unlikely phenomenon has ever been
> observed to occur . . .

It seems to me that mountains often are found in clusters. That islands
are often found in clusters. And those are the metaphors we are using
for beneficial states. They (mountains, islands, and biologically
useful activities) occur in clusters because of causal reasons, not
random ones.

>>You cannot simply assume an "average"
>>distribution in the absence of background information: you have to find
>>out precisely the kind of distribution you are dealing with. And even
>>if you do find that the distribution is "stacked", it does not imply
>>that an intelligence was involved.
>
>
> Oh really? You think that stacking the deck as you have done can
> happen mindlessly in less than zillions of years of average time?
> Come on now! What planet are you from?

When you start with useful rather than random sequences in a
pre-existing organism, you are necessarily stacking the deck in a search
for other related useful sequences. Especially if the search were not
random (but followed the simple rules I gave to my blind man), did not
occur on a perfectly flat plane, and did not start with a search from
one random site but from many non-random partially useful sites. Only
the ones that *start* off close to the desired island/mountain have a
good chance of reaching a useful end point, but that is merely probability.

>>The stacking could occur due to the
>>constraints imposed by the very definition of the problem: in the case
>>of evolutions, by the physical constraints governing the interactions
>>between the molecules involved in biological systems.
>
>
> Oh, so the physical laws of atoms and molecules force them to
> self-assemble themselves in functionally complex systems?

As a matter of fact, it is indeed the physical laws of atoms and
molecules that cause the self-assembly of structures like flagella from
their component parts. There is no intelligent assembler of flagella in
bacteria. You keep confusing and confounding the self-assembly of
flagella (or ribosomes, or cilia, or mitochondrial spindles) in cells
with their evolutionary points of origin. Please use these terms correctly.

Just so you know, I suspect he was talking about the constraints
involved in the evolution, say, of a glycoside hydrolysis. One of these
constraints being the ability to bind a specific glycoside. This
probably requires the presence of a binding cleft in the protein, thus
limiting the evolution of beta galactosidases to modifications of
molecules that have a cleft capable of binding the sugar galactose
linked through a betagalactoside linkage to another molecule. For
example, ebg or immunoglobins (yep, that cleft can be modified to make
an immunoglobin an effective lactase). The hard part in evolving a
lactase from an immunoglobin is in having the right few amino acids
needed to weaken the bond to be hydrolyzed and in not having binding be
so tight that the products are not released.

> Now you are
> really reaching. Tell me why the physical constraints of these
> molecular machines force all beneficial possibilities to be so close
> together? This is simply the most ludicrous notion that I have heard
> in a very long time. You would really do well in Vegas with that one!
> Try telling them, when they come to arrest you for cheating, that the
> deck was stacked because of the physical constraints of the playing
> cards.

The above makes no sense at all as written when compared to reality. I
suspect that Sean misunderstood what Robin meant. Surely Sean must
realize that all the complex structures in cells self-assemble in these
cells because of simple chemical and physical affinities. There are no
little homunculi working on assembly lines in cells, willing to go on
strike for higher wages (MORE ATP!), etc. That would be carrying the
idea of intelligence involved in these processes a step too far.

>>In fact, why
>>would you expect that the regular and highly predictable physical laws
>>governing biochemical reactions would produce a random, "average"
>>distribution of "beneficial sequences"?
>

I wouldn't expect new beneficial sequences to be random. I would expect
new "beneficial sequences" to be close to one or more of the
pre-existing "beneficial sequences" in a cell. That is because the
'new' needs of a cell are most often going to involve molecules with
similarity to molecules that are already biologically relevant. That
is, I suspect that there will be clusters of 'beneficial' sequences.
Why do think 'new' beneficial sequences are evenly spaced throughout
sequence space, but always very, very far away from any current sequence?

>
> Because, I don't know of any requirement for them to be clustered
> outside of deliberate design - do you? I can see nothing special
> about the building blocks that make up living things that would cause
> the potentially beneficial systems found in living things to have to
> be clustered (just like there is nothing inherent in playing cards
> that would cause them to stack themselves in any particular order).

I *do* expect to see clustering in useful sequences. And I *do* see it.
One regularly sees families of genes rather than genes with no
sequence similarity. For example, a big chunk of genes are very similar
as membrane-spanning proteins, but differ in the allosteric effector
that transduces an effect across the membrane in eucaryotes. I expect
to see things like the similarity in the TTSS proteins and flagellar
proteins rather than seeing completely different proteins. The reason I
*do* expect to see such clustering is because I think these features
arose by descent with modification rather than by a random walk from a
random starting point to an end that is unrelated to the starting point.
The reason I *do* see such clustering is because descent with
modification is how nature works to produce new proteins. The reason I
don't see complete randomness in new sequence is because your model of
evolution is a bogus strawman.

> However, if you know of a reason why the physical nature of the
> building blocks of life would force them to cluster together despite
> having a low density in sequence space, please, do share it with me.

Sequences of utility cluster together because they arose by common
descent and descent with modification rather than by random walks
through random sequence space from a random starting point.

> Certainly none of your computer examples have been able to demonstrate
> such a necessity. Why then would you expect such a forced clustering
> in the potentially beneficial states of living things?

Look at an evolutionary branching tree. You will see clustering of
exactly the type one sees in sequences. Not *just* similar. Exactly.

>>>>For an extreme
>>>>example, consider a space of strings consisting of length 1000, where
>>>>each position can be occupied by one of 10 possible characters.
>>
>>Note, I wrote, "extereme example". My point was *not* invent a
>>distribution which makes it likely for evolutiuon to occur (this example
>>has about as much to do with evolution as ballet does with quantum
>>mechanics), but to show how inadequate your methods are.
>
>
> Actually, this situation has a lot to do with evolution and is the
> real reason why evolution is such a ludicrous idea.

No, Sean. It has a lot to do with your bogus straw man of evolution.
It has nothing to do with reality.

> What your
> illustration shows is that only if the deck is stacked in a most
> unlikely way will evolution have the remotest possibility of working.
> That is what I am trying to show and you demonstrated this very
> nicely. Unwittingly it is you who effectively show just how
> inadequate evolutionary methods are at making much of anything outside
> of an intelligently designed stacking of the deck.

[Snip much more of little interest, since GIGO is GIGO whether it is
done in one paragraph or twenty]

Sean Pitman

unread,
Jan 14, 2004, 3:52:22 PM1/14/04
to
"Chris Merli" <clm...@insightbb.com> wrote in message news:<GTcNb.65427$xy6.124383@attbi_s02>...

> >
> > Consider the scenario where there are 10 ice cream cones on the
> > continental USA. The goal is for a blind man to find as many as he
> > can in a million years. It seems that what you are suggesting is that
> > the blind man should expect that the ice cream cones will all be
> > clustered together and that this cluster will be with arms reach of
> > where he happens to start his search. This is simply a ludicrous
> > notion outside of intelligent design. My hypothesis, on the other
> > hand, suggests that these 10 ice cream cones will have a more random
> > distribution with hundreds of miles separating each one, on average.
> > An average starting point of the blind man may, by a marvelous stroke
> > of luck, place him right beside one of the 10 cones. However, after
> > finding this first cone, how long, on average, will it take him to
> > find any of the other 9 cones? That is the question here. The very
> > low density of ice cream cones translates into a marked increase in
> > the average time required to find them. Now, if there were billions
> > upon billions of ice cream cones all stuffed into this same area, then
> > one could reasonably expect that they would be separated by a much
> > closer average distance - say just a couple of feet. With such a high
> > density, the average time needed for the blind man to find another ice
> > cream cone would be just a few seconds.
>
> But there is not one blind man looking there are many and only those close
> enough to the cluster of cones in the first place are likely to succeed.

Exactly right. The problem is that increasing the number of blind men
searching only helps for a while, at the lowest levels of functional
complexity where the density of ice cream cones is the greatest.
However, with each step up the ladder of functional complexity, the
density of ice cream cones decreases in an exponential manner. In
order to keep up with this exponential decrease in average cone
density, the number of blind men has to increase exponentially in
order to find the rarer cones at the same rate. Very soon the
environment cannot support any more blind men and so they must
individually search out exponentially more and more sequence space, on
average, before success can be realized (i.e., a cone or cluster of
cones is found). For example, it can be visualized as stacked levels
of rooms. Each room has its own average density of ice cream cones.
The rooms on the lowest level have the highest density of ice cream
cones - say one cone every meter or so, on average. Moving up to the
next higher room the density decreases so that there is a cone every 2
meters or so. Then, in the next higher room, the density decreases to
a cone every 4 meters or so, on average. And, it goes from there.
After 30 or so steps up to higher levels, the cone density is 1 every
billion meters or so, on average.

Are you starting to see the problem? What one blind man could find in
just a few seconds at the lowest levels, thousands of blind men cannot
find in thousands of years after just a few step up into the higher
levels. Clustering doesn't help them out here. Because, on average,
the blind men just will not happen to start out close to a cluster of
cones. And, if they do happen to get so fortunate as to end up close
to a rare cluster, what are the odds that they will find another
cluster of cones within that same level? You must think about the
*average* time involved, not the unlikely scenario that finding one
cluster solves all problems. Clustering, contrary to what many have
suggested, does not increase the average density of beneficial states
at a particular level of sequence space. This means that clustering
does not decrease the average time required to find a new ice cream
cone. In fact, if anything, clustering would increase the average
time required to find a new ice cream cone.

> > > You cannot simply assume an "average"
> > > distribution in the absence of background information: you have to find
> > > out precisely the kind of distribution you are dealing with. And even
> > > if you do find that the distribution is "stacked", it does not imply
> > > that an intelligence was involved.
> >
> > Oh really? You think that stacking the deck as you have done can
> > happen mindlessly in less than zillions of years of average time?
> > Come on now! What planet are you from?
>
> Lets talk clusters. How many point mutations of a protein are in fact still
> functional. This tends to create a cluster all of its own. Given this fact

> the idea that they are spread evenly across the landscape is just not true.

Certainly the various beneficial functions are indeed clustered. But
you must realize that clustering doesn't help you find a new cluster
with a new type of function any faster. Say that you start on a
particular clustered island of function. You can move around this
island pretty easily. But, the entire island pretty much does the
same type of function. The question is, how long will it take, on
average, to find a new island of states/sequences with a new type of
function? In order to solve this problem you must have some idea
about the *average* density of all beneficial states in sequence space
as they compare to the non-beneficial sequences that also exist in
sequence space. This average density will tell you, clustered or not,
how long it will take to find a new sequence with a new type of
function via random walk across the non-beneficial sequences. If
fact, the more clustered the sequences are, the longer it will take,
on average to find a new cluster.

Of course Robin, Howard, and many others in this forum have tried to
float the idea that these islands will all happen to be clustered
neatly around the starting point by some unknown but necessary force
of nature despite incredibly low average densities given the overall
volume of sequence space at that level of complexity. They are
basically suggesting that evolution works because the deck is stacked
neatly in favor of evolutionary processes. Of course, for evolution
to really work such deck stacking would not only be helpful, but
vital. Evolution simply cannot work unless the deck is marvelously
stacked in its favor like this. But, what are the odds that the deck
would be so neatly stacked like this outside of intelligent design?
That is the real question here. And so far, no evolutionist that I
have yet encountered seems to be able to answer this question in a way
that makes any sort of rational sense to me. Perhaps you are better
able to understand the solution to this problem than I am?

Sean
www.naturalselection.0catch.com

Frank J

unread,
Jan 14, 2004, 7:13:27 PM1/14/04
to
"\"Rev Dr\" Lenny Flank" <lflank...@ij.net> wrote in message news:<3ff86071$1...@corp.newsgroups.com>...

> Sean Pitman wrote:
>
>
> >
> > Until then, this is all I have time for today.
>
>
> Hey doc, when will you have time to tell us what the scientific theory
> of intelligent design is --- what does the desigher do, specifically,
> what mechanisms does it use to do it, where can we see these mechanisms
> in operation today. And what idnicates there is only one desinger and
> not, say, ten or fifty of them all working together.


C'mon, one question at a time. And good luck getting any answer since
I am still waiting for him and several others to answer my simple
question to define "common design."

>
> After that, can you find the time to explain to me how ID "theory" is
> any less "materialist" or "naturalist" or "atheist" than is evolutionary
> biology, since ID "theory" not only does NOT hypothesize the existence
> of any supernatural entities or actions, but specifically states that
> the "intelligent designer" might be nothing but a space alien.


ID may be less "naturalistic," but only because it rarely makes
testable claims to support its own model. But when it does, it is
every bit as "naturalistic" as evolution and the
mutually-contradictory creationisms. Too bad those claims fail every
time.

And ID and creationism are no less "atheistic" than evolution,
because, as you know, and as anti-evolutionists don't want anyone to
know, evolution never specifically rules out an "intelligent
designer." Ironically it is the anti-evolutionists who constantly
promote "atheistic science" by their false dichotomy.


>
> And after THAT, could you find the time to tell us how you apply
> anything other than "naturalism" or "materialism" to your medical
> practice? What non-naturalistic cures do you recommend for your
> patients, doctor.

My guess is that he says: "Oo ee oo ah ah ting tang walla walla bing
bang."

Chris Merli

unread,
Jan 14, 2004, 9:13:16 PM1/14/04
to

"Sean Pitman" <seanpi...@naturalselection.0catch.com> wrote in message
news:80d0c26f.0401...@posting.google.com...

This is based on the false assumption that increasing complexity must entail
de novo development of the more complex systems. It is painfully clear from
an examination of most proteins that even within a single polypeptide there
are portions that are recruited from other coding sequences. Thus the basic
units that even you have realized can evolve are easily shuffled copied and
adapted. I would contend in fact that the hardest part of the evolution is
not the complex systems that you have argued but the very simple functions.

In
> order to keep up with this exponential decrease in average cone
> density, the number of blind men has to increase exponentially in
> order to find the rarer cones at the same rate. Very soon the
> environment cannot support any more blind men and so they must
> individually search out exponentially more and more sequence space, on
> average, before success can be realized (i.e., a cone or cluster of
> cones is found). For example, it can be visualized as stacked levels
> of rooms. Each room has its own average density of ice cream cones.
> The rooms on the lowest level have the highest density of ice cream
> cones - say one cone every meter or so, on average. Moving up to the
> next higher room the density decreases so that there is a cone every 2
> meters or so. Then, in the next higher room, the density decreases to
> a cone every 4 meters or so, on average. And, it goes from there.
> After 30 or so steps up to higher levels, the cone density is 1 every
> billion meters or so, on average.

If the development of each protein started from scratch you may have an
excellent arguement but nearly all proteins from other proteins so you are
starting from a point that is known to be functional.

Have you ever really considered how many functions proteins provide. At the
very basic level there are very few. All those complex functions are based
on only a few very simple things that can occur at a link between two amino
acids plus some chemical and electrical forces. Look at the active site of
most enzymes and you will find them remarkably simple.

I am afraid I will simply have to wait for evidence to elucidate the reason
for this. I asked you before what evidence you had that these clusters do
not exist and based on your reply here it is safe to assume the answer is
none. Not only do you not know if there is clustering but you are not even
certain what percentage of the protein sequences are functional in any way.
Based on this it is very hard to lend any weight to your speculations.
Could you present a experiment that would support any of your assumptions?
Please do not present experiments that would require negative results as
those are not scientific.

>
> Sean
> www.naturalselection.0catch.com
>

Sean Pitman

unread,
Jan 14, 2004, 9:21:03 PM1/14/04
to
howard hershey <hers...@indiana.edu> wrote in message news:<bu46sv$srt$1...@hood.uits.indiana.edu>...


> > Consider the scenario where there are 10 ice cream cones on the
> > continental USA. The goal is for a blind man to find as many as he
> > can in a million years.
>
> Except that is NOT what evolution does. Evolution starts with an
> organism with pre-existing sequences that produce products and interact
> with environmental chemicals in ways that are useful to the organism's
> reproduction.

Yes . . . so start the blind man off with an ice-cream cone to begin
with and then have him find another one.

> The situation is more like 10,000 blind men in a varying
> topography who blindly follow simple and dumb rules of the game to find
> useful things (ice cream at the tops of fitness peaks):

You don't understand. In this scenario, the positively selectable
topography is the ice-cream cone. There are no other selectable
fitness peaks here. The rest of the landscape is neutral. Some of
the ice-cream cones may be more positively selectable than others
(i.e., perhaps the man likes vanilla more than chocolate). However,
all positive peaks are represented in this case by an ice-cream cone.

> Up is good. Down
> is bad.

Ice-cream cone = Good or "Up" (to one degree or another) or even
neutral depending upon one's current position as it compares to one's
previous position. For example, once you have an ice cream, that is
good. But, all changes that maintain that ice cream but do not gain
another ice cream are neutral.

No ice-cream cone = "Bad", "Down", or even "neutral" depending upon
one's current position as it compares to one's previous position.

> Flat is neither good nor bad.

Exactly. Flat is neutral. The more neutral space between each "good"
upslope/ice-cream cone, the longer the random walk. The average
distance between each selectable "good" state translates into the
average time required to find such a selectable state/ice-cream cone.
More blind men searching, like 10,000 of them, would cover the area
almost 10,000 times faster than just one blind man searching alone.
However, at increasing levels of complexity the flat area expands at
an exponential rate. In order to keep up and find new functions at
these higher levels of functional complexity, the population of blind
men will have to increase at an equivalent rate. The only problem
with increasing the population is that very soon the local environment
will not be able to support any larger of a population. So, if the
environment limits the number of blind men possible to 10,000 - that's
great if the average neutral distance between ice-cream cones in a few
miles or so, but what happens when, with a few steps up the ladder of
functional complexity, the neutral distance expands to a few trillion
miles between each cone, on average? Now each one of your 10,000
blind men have to search around 50 million sq. miles, on average,
before the next ice-cream cone or a new cluster of ice cream cones
will be found by even one blind man in this population.

> Keep walking in all cases.

They keep walking alright - a very long ways indeed before they reach
anything beneficially selectable at anything very far beyond the
lowest levels of functional complexity.

> It
> would not take too long for these 10,000 blind men to be found in
> decidedly non-random places (the high mesas of functional utility where
> they are wandering around the flat tops if you haven't guessed).

There is a funny thing about these mesas. At low levels of
complexity, these mesas are not very large. In fact, many of them are
downright tiny - just one or two steps wide in any direction and a
new, higher mesa can be reached. However, once a blind man finds this
new mesa new higher mesa (representing a different type of function at
higher level of specified complexity) and climbs up onto its higher
surface, the distance to a new mesa at the same height or taller is
exponentially greater than it was at the lower levels of mesas.

___
__ __
_ _-_ __
__-_
_-_- -__-__-_- _-__-_-_-__- -_- _-_-_
_-_-__

> And
> the ice cream cones (the useful functions), remember, are not randomly
> distributed either. They are specifically at the tops of these mesas as
> well. That is what a fitness landscape looks like.

Actually, the mesa itself, every part of its surface, represents an
ice cream cone. There is no gradual increase here. Either you have
the ice-cream cone or you don't. If you don't have one that is even
slightly "good"/beneficial, then you are not higher than you were to
begin with and you must continue your random walk on top of the flat
mesa that you first started on (i.e., your initial beneficial
function(s)).

> If this topography of utility only changed slowly, at any given time it
> would appear utterly amazing to Sean that the blind men will all be
> found at these local high points or optimal states (the mesas licking
> the ice cream cones on them) rather than being randomly scattered around
> the entire surface.


If all the 10,000 blind men started at the same place, on the same
point of the same mesa, and then went out blindly trying to find a
higher mesa than the one they started on, the number that they found
would be directly proportional to the average distance between these
taller mesas. If the density of taller mesas, as compared to the one
they are now on, happens to be say, one every 100 meters, then they
will indeed find a great many of these in short order. However, if
the average density of taller mesas, happens to be one every 10,000
kilometers, then it would take a lot longer time to find the same
number of different mesas as compared to the number the blind men
found the first time when the mesas were just 100 meters apart.

> They reached these high points (with the ice cream)
> by following a simple dumb algorithm.

Yes - and this mindless "dumb" algorithm works just fine to find new
and higher mesas if and only there is a large average density of mesas
per given unit of area (i.e., sequence space). That is why it is easy
to evolve between 3-letter sequences. The ratio/density of such
sequences is as high as 1 in 15. Any one mutating sequence will find
a new 3-letter sequence within 15 random walk steps on average. A
population of 10,000 such sequences (blind men) would find most if not
all the beneficial 3-letter words (ice-cream cones) in 3-letter
sequence space in less than 30 generations (given that there was one
step each, on average, per generation).

This looks good so far now doesn't it? However, the problems come as
you move up the ladder of specified complexity. Using language as an
illustration again, it is not so easy to evolve new beneficial
sequences that require say, 20 fairly specified letters, to transmit
an idea/function. Now, each member of our 10,000 blind men is going
to have to take over a trillion steps before success (the finding of a
new type of beneficial state/ice cream cone) is realized for just one
of them at this level of complexity.

Are we starting to see the problem here? Of course, you say that
knowledge about the average density of beneficial sequences is
irrelevant to the problem, but it is not irrelevant unless you, like
Robin, want to believe that all the various ice-cream cones
spontaneously cluster themselves into one tiny corner of the potential
sequence space AND that this corner of sequence space just so happens
to be the same corner that your blind men just happen to be standing
in when they start their search. What an amazing stroke of luck that
would be now wouldn't it?

> But you were wondering how something new could arise *after* the blind
> men are already wandering around the mesas? The answer is that it
> depends. They can't always do so.

And why not Howard? Why can't they always do so? What would limit
the blind men from finding new mesas? I mean really, each blind man
will self-replicate (hermaphrodite blind men) and make 10,000 new
blind men on the mesa that he/she/it now finds himself on. This new
population would surely be able to find new mesas in short order if
things worked as you suggest. But the problem is that if the mesas
are not as close together, on average, as they were at the lower level
where the blind men first started their search, it is going to take
longer time to find new mesas at the same level or higher. That is
the only reason why these blind men "can't always" find "something
new". It has to do with the average density of mesas at that level.

> But remember that these pre-existing
> mesas are not random places. They do something specific with local
> utility.

The mesas represent sequences with specific utilities. These
sequences may in fact be widely separated mesas even if they happen to
do something very similar. Really, the there is no reason for the
mesas to be clustered in one corner of sequence space. A much more
likely scenario is for them to be more evenly distributed throughout
the potential sequence space. Certainly there may be clusters of
mesas here and there, but on average, there will still be a wide
distribution of mesas and clusters of mesas throughout sequence space
at any given level. And, regardless of if the mesas are more
clustered or less clustered, the *average* distance between what is
currently available and the next higher mesa will not be significantly
affected.

> Let's say that each mesa top has a different basic *flavor* of
> ice cream. Say that chocolate is a glycoside hydrolase that binds a
> glucose-based glycoside. Now let's say that the environment changes so
> that one no longer needs this glucose-based glycoside (the mesa sinks
> down to the mean level) but now one needs a galactose-based glycoside
> hydrolase.

You have several problems here with your illustration. First off,
both of these functions are very similar in type and use very similar
sequences. Also, their level of functional complexity is relatively
low (like the 4 or 5 letter word level). Also, you must consider the
likelihood that the environment would change so neat so that galactose
would come just when glucose is leaving. Certainly if you could
program the environment just right, in perfect sequence, evolution
would be no problem. But you must consider the likelihood that the
environment will change in just the right way to make the next step in
an evolutionary sequence beneficial when it wasn't before. The odds
that such changes will happen in just the right way on both the
molecular level and environmental level get exponentially lower and
lower with each step up the ladder of functional complexity. What was
so easy to evolve with functions requiring no more than a few hundred
fairly specified amino acids at minimum, is much much more difficult
to do when the level of specified complexity requires just a few
thousand amino acids at minimum. It's the difference between evolving
between 3-letter words and evolving between 20-letter phrases. What
are the odds that one 20-letter phrase/mesa that worked well in one
situation will sink down with a change in situations to be replaced by
a new phrase of equal complexity that is actually beneficial? -
Outside of intelligent design? That is the real question here.

> Notice that the difference in need here is something more
> like wanting chocolate with almonds than wanting even strawberry, much
> less jalapeno or anchovy-flavored ice cream. The blind man on the newly
> sunk mesa must keep walking, of course, but he is not thousands of miles
> away from the newly risen mesa with chocolate with almonds ice cream on
> top.

He certainly may be extremely far away from the chocolate with almonds
as well as every other new type of potentially beneficial ice cream
depending upon the level of complexity that he happens to be at (i.e.,
the average density of ice-creams of any type in the sequence space at
that level of complexity).

> Changing from one glucose-based glycoside hydrolase to one with a
> slightly different structure is not the same as going from chocolate to
> jalapeno or fish-flavored ice cream. Not even the same as going from
> chocolate to coffee. The "island" of chocolate with almonds is *not*
> going to be way across the ocean from the "island" of chocolate.

Ok, lets say, for arguments sake, that the average density of
ice-cream cones in a space of 1 million square miles is 1 cone per 100
square miles. Now, it just so happens that many of the cones are
clustered together. There is the chocolate cluster with all the
various types of chocolate cones all fairly close together. Then,
there is the strawberry cones with all the variations on the
strawberry theme pretty close together. Then, there is the . . .
well, you get the point. The question is, does this clustering of
certain types of ice creams help is the traversing the gap between
these clustered types of ice creams? No it doesn't. If anything, the
clustering only makes the average gap between clusters wider. The
question is, how to get from chocolate to strawberry or any other
island cluster of ice creams when the average gap is still quite
significant?

You see, the overall average density of cones is still significant to
the problem no matter how you look at it. Clustering some of them
together is not going to help you find the other clusters - unless
absolutely all of the ice cream islands are clustered together as well
in a cluster of clusters all in one tiny portion of the overall
potential space. This is what Robin is trying to propose, but I'm
sorry, this is an absolutely insane argument outside of intelligent
design. How is this clustering of clusters explained via mindless
processes alone?

> It will
> be nearby where the blind man is. *And* because chocolate with almonds
> is now the need, it will also be on the new local high mesa (relative to
> the position of the blind man on the chocolate mesa). The blind man
> need only follow the simple rules (Up good. Down bad. Neutral neutral.
> Keep walking.) and he has a good chance of reach the 'new' local mesa
> top quite often.

And what about the other clusters? Is the environment going to change
just right a zillion times in a row so that bridges can be built to
the other clusters?

> And remember that there is not just one blind man on one mesa in this
> ocean of possible sequences. There are 10,000 already present on 10,000
> different local mesas with even more flavors than the 31 that most ice
> cream stores offer. Your math always presuposes that whenever you need
> to find, say, vanilla with cherry the one blind man starts in some
> random site and walks in a completely random fashion (rather than by the
> rules I pointed out) across half the universe of sequence space to reach
> your pre-determined goal by pure dumb luck to find the perfect lick.

That is not my position at all as I have pointed out to you numerous
times. It seems that no matter how often I correct you on this straw
man caricature of my position you make the same straw man assertions.
Oh well, here it goes again.

I'm perfectly fine with the idea that there is not just one man, but
10,000 or many more men already in place on different mesas that are
in fact selectably beneficial. In fact, there may be 10,000 or more
men on each of 10,000 mesas. That is all perfectly fine and happens
in real life. When something new "needs to be found", say, "vanilla
with a cherry on top" or any other potentially beneficial function at
that level of complexity or greater (this is not a teleological search
you know since there are many ice-cream cones available), all of the
men may search at the same time.

My math certainly does not and never did presuppose that only one man
may search the sequence space. That is simply ridiculous. All the
men search at the same time (millions and even hundreds of billions of
them at times). The beneficial sequences are those sequences that are
even slightly better than what is currently had by even one member of
the vast population of blind men that is searching for something new
and good.

Now, if the average density of something new and good that is even
slightly selectable as new and good is less than 1 in a trillion
trillion, even 100 billion men searching at the same time will take a
while to find something, anything, that is even a little bit new and
good at the same level of specified complexity that they started with.
On average, none of the men on their various mesas will be very close
to any one of the new and good mesas within the same or higher levels
of sequence space if the starting point is very far beyond the lowest
levels of specified complexity.

> My
> presumption is that the successful search is almost always going to
> start from the pre-existing mesa

Agreed.

> with the closest flavor to the new need
> (or from a duplicate, which, as a duplicate, is often superfluous and
> quickly erodes to ground level in terms of its utility).

This is where we differ. Say you have chocolate and vanilla. Getting
to the different varieties of chocolate and vanilla is not going to be
much of a problem. But, say that neither chocolate nor vanilla are
very close to strawberry or to each other. Each cluster is separated
from the other clusters by thousands of miles. Now, even though you
already have two clusters in your population, how are you going to
evolve the strawberry cluster if an environmental need arises where it
would be beneficial?

You see, you make the assumption that just because you start out with
a lot of clusters that any new potentially beneficial sequence or
cluster of sequences will be fairly close to at least one of your
10,000 starting clusters. This is an error when you start considering
levels of sequence space that have very low overall densities of
beneficial sequences. No matter where you start from and no matter
how many starting positions you have to begin with, odds are that the
vast majority of new islands of beneficial sequences will be very far
away from everything that you have to start with beyond the lowest
levels of functional complexity.

> As mentioned,
> these pre-existing mesas are not random pop-ups. They are at the most
> useful places in sequence space from which to try to find near-by mesas
> with closely-related biologically useful properties because they already
> have biologically useful properties.

Yes, similar useful biological properties would all be clustered
together under one type of functional island of sequences. However,
the overall density of beneficial sequences in sequence space dictates
how far apart, on averages, these clusters of clusters will be from
each other. New types of functions that are not so closely related
will most certainly be very far away from anything that you have to
start with beyond the lowest levels of functional complexity. You may
do fine with chocolate and vanilla variations since those are what you
started with, but you will have great difficulty finding anything
else, such as strawberry, mocha, caviar, etc . . .

The suggestion that absolutely all of the clusters are themselves
clustered together in a larger cluster or archipelago of clusters in a
tiny part of sequence space is simply a ludicrous notion to me -
outside of intelligent design that is. Oh no, you, Robin, Deaddog,
Sweetness, Musgrave, and all the rest will have to do a much better
job and explaining how all the clusters can get clustered together
(when they obviously aren't) outside of intelligent design.



> I *do* expect to see clustering in useful sequences. And I *do* see it.

So do I. Who is arguing against this? Useful sequences are often
clustered around a certain type of function. What I am talking about
is evolution between different types of functions. The evolution of
different sequences with the same basic type of function is not an
issue at all. It happens all the time, usually in the form of an
up-regulation or down-regulation of a certain type of function, even
at the highest levels of functional complexity. But, this sort of
intra-island evolution is a far cry from evolving a new type of
function (i.e., going from one cluster to another). In fact, this
sort of evolution never happens beyond the lowest levels of functional
complexity due to the lack of density of beneficial sequences at these
higher levels of specified complexity.

In any case, this is all I have time for today. As always, it has
been most interesting. Please do try again . . .

Sean
www.naturalselection.0catch.com

Jethro Gulner

unread,
Jan 15, 2004, 12:17:26 AM1/15/04
to
I'm thinking TSS to flagellum is on the order of chocolate to
chocolate-fudge-brownie

howard hershey <hers...@indiana.edu> wrote in message news:<bu46sv$srt$1...@hood.uits.indiana.edu>...

david ford

unread,
Jan 15, 2004, 8:51:08 AM1/15/04
to
Sean Pitman <seanpi...@naturalselection.0catch.com> on 4 Jan 2004:
RobinGoodfellow <lmuc...@yahoo.com>:

[snip]

There's no such thing as an intelligent designer.
What I meant to say was, there's no such thing as an intelligent
designer of computer programs.
What I really meant to say was, there's no such thing as an
intelligent designer(s) of biology.

> If
> enough sequences are defined as beneficial and they are placed in just
> the right way, with the right number of spaces between them, then
> certainly such a high ratio will result in rapid evolution - as we saw
> here. However, when neutral non-defined gaps are present, they are a
> real problem for evolution. In this case, a gap of just 16 neutral
> mutations effectively blocked the evolution of the EQU function.
>
> http://naturalselection.0catch.com/Files/computerevolution.html

[snip]

>> The answer is simple - the ratio of beneficial states does NOT
matter!
>
> Yes it does. You are ignoring the highly unlikely nature of your
> scenario. Tell me, how often do you suppose your start point would
> just happen to be so close to the only other beneficial sequence in
> such a huge sequence space? Hmmmm? I find it just extraordinary that
> you would even suggest such a thing as "likely" with all sincerity of
> belief. The ratio of beneficial to non-beneficial in your
> hypothetical scenario is absolutely miniscule and yet you still have
> this amazing faith that the starting point will most likely be close
> to the only other "winning" sequence in an absolutely enormous
> sequence space?! Your logic here is truly mysterious and your faith
> is most impressive.

Anything is possible with enough faith. Simply believe hard enough,
and reality _will_ conform.

> I'm sorry, but I just can't get into that boat
> with you. You are simply beyond me.

What are you afraid of-- getting a little wet? When the boat sinks,
you will, after all, be able to swim. Though I don't know for how
long....



>> All that matters is their distribution, and how well a particular
>> random walk is suited to explore this distribution.
>
> Again, you must consider the odds that your "distribution" will be so
> fortuitous as you seem to believe it will be. In fact, it has to be
> this fortuitous in order to work. It basically has to be a set up for
> success. The deck must be stacked in an extraordinary way in your
> favor in order for your position to be tenable. If such a stacked
> deck happened at your table in Las Vegas you would be asked to leave
> the casino in short order or be arrested for "cheating" by intelligent
> design since such deck stacking only happens via intelligent design.

Intelligent design advocates often cheat. They are masters of
illusion and sleight of hand. Their ideas adapt to data as fog adapts
to land, to borrow some phraseology from the creationist ReMine.
Their views can "explain" any conceivable observation, and any
conceivable set of circumstances (exception: if biology did not
exist, or if we are living in the Matrix, and what we think is real is
not real and is a dream).

Magician Walter ReMine wrote the extremely dangerous and execrable
book _The Biotic Message: Evolution versus Message Theory_ (1993),
538pp. I cannot urge upon you strongly enough the importance of not
reading that book. Miraculously, my faith in the solidity and rigor
of the theory of evolution aka Precious survived the reading of large
portions of that most despicable book. Those were dark times, but my
faith in Precious survived.

[snip]

>> A random walk
>> starting at a given beneficial sequence, and allowing certain
>> transitions from one sequence to another, would require a
completely
>> different type of analysis. In the analyses of most such search
>> algorithms, the "ratio" of beneficial sequences would be irrelevant
-
>> it is their *distribution* that would determine how well such an
>> algorithm would perform.
>
> The most likely distribution of beneficial sequences is determined by
> their density/ratio. You cannot simply assume that the deck will be
> so fantastically stacked in the favor of your neat little evolutionary
> scenario. I mean really, if the deck was stacked like this with lots
> of beneficial sequences neatly clustered around your starting point,
> evolution would happen very quickly. Of course, there have been those
> who propose the "Baby Bear Hypothesis". That is, the clustering is
> "just right" so that the theory of evolution works.

How could the existence of such just-right clustering be accounted
for-- what could have produced it?
In your response, please do not invoke intelligence. After all, in
the story of Goldilocks and the three bears, the porridge was not
prepared by intelligence. Intelligence cannot account for the
appearance of _anything_. This post is an illustration of that fact.

> That is the best of

Sorry, not interested. I recently blew my life savings on 50 acres of
oceanfront property on the Moon.
If you act now, you too can get in on this amazing ground-level deal,
and be privy to the secrets of the Oceanfront Moon Property Society.
The only requirements for membership are that you own Moon property
and affirm that intelligence/ mind cannot be an explanation for the
appearance of anything, especially biology.

Sean Pitman

unread,
Jan 15, 2004, 11:30:52 AM1/15/04
to
jethro...@bigfoot.com (Jethro Gulner) wrote in message news:<edf04d4a.04011...@posting.google.com>...

>
> I'm thinking TSS to flagellum is on the order of chocolate to
> chocolate-fudge-brownie

Now that's a serious stretch of the imagination. The TTSS system is a
non-motile secretory system while the fully formed flagellar system is
a motility system as well. The TTSS system requires 6 or so different
protein parts, at minimum, for its formation while the motility
function of the flagellar system requires and additional 14 or so
different protein parts (for a total of over 20 parts) before its
motility function can be realized. Unless you can find intermediate
functions for the gap of more than a dozen required parts that
separate the TTSS system from the Flagellar system, I'd say this gap
is quite significant indeed, requiring at minimum several thousand
fairly specified amino acids. Certainly this is not the same thing as
roaming around the same island cluster with the same type of function.
The evolution form the TTSS island of function to the brand new type
of motility function found in the flagellar island would have to cross
a significant distance before the motility function of the flagellum
could be realized. Such a distance could not be crossed via random
walk alone this side of zillions of years in any population of
bacteria on Earth. In order for evolution to have truly crossed such
a gap, without intelligent design helping it along, there would have
to be a series of closely spaced beneficial functions/sequences
between the TTSS and the motility function of the flagellum.

Where is this series of steppingstones? That is the real question!
Many have tried to propose the existence of various stepping-stone
functions, but none have been able to show that these steppingstones
could actually work as no one has ever shown the crossing from any
proposed steppingstone to any other in real life. If you think you
know better how such a series could exist and actually work to
eliminate this gap problem, please do share your evolutionary sequence
with us.

Sean
www.naturalselection.0catch.com

Sean Pitman

unread,
Jan 15, 2004, 2:50:33 PM1/15/04
to
"Chris Merli" <clm...@insightbb.com> wrote in message news:<lKmNb.55023$5V2.67607@attbi_s53>...

> >
> > Exactly right. The problem is that increasing the number of blind men
> > searching only helps for a while, at the lowest levels of functional
> > complexity where the density of ice cream cones is the greatest.
> > However, with each step up the ladder of functional complexity, the
> > density of ice cream cones decreases in an exponential manner.
>
> This is based on the false assumption that increasing complexity must entail
> de novo development of the more complex systems. It is painfully clear from
> an examination of most proteins that even within a single polypeptide there
> are portions that are recruited from other coding sequences. Thus the basic
> units that even you have realized can evolve are easily shuffled copied and
> adapted. I would contend in fact that the hardest part of the evolution is
> not the complex systems that you have argued but the very simple functions.

This is a very common misconception among evolutionists - that if the
right subparts of a system are similar or identical to other parts
elsewhere in other systems, that the system in question obviously
arose via a "simple" assembly of pre-existing subparts.

The problem with this idea is that just because all of the right
subparts needed to make a new beneficial system of function are there,
already fully formed as parts of other systems of function, does not
mean that they will self-assemble themselves to form a new collective
system of function. For example, all of the individual amino acids
are there, fully formed, to make a motility apparatus in a
historically non-motile bacterial colony. Say that motility would be
advantageous to this colony if it evolved a system that would give it
motility. All the right parts are there, but they don't know how to
self-assemble themselves to make such a system.

Now why is this? Because, in order for correct assembly of the parts
to proceed, the information for their assembly must be pre-established
in the DNA. This genetic information tells where, when, and how much
of each part to make so that the assembly of the molecular systems can
occur. Without this pre-established information the right parts just
won't assembly properly beyond the lowest levels of functional
complexity. It would be like having all the parts to a watch in a
bag, shaking the bag for a billion years, and expecting a fully formed
watch, or anything else of equal or greater emergent functional
complexity, to fall out at the end of that time. The same is true for
say, a bacterial flagellum. Take all of the necessary subparts needed
to make a flagellum, put them together randomly, and see if they will
self-assemble a flagellar apparatus. It just doesn't happen outside
of the very specific production constraints provided by the
pre-established genetic information that code for both flagellar part
production as well as where, when, and how much part to produce so
that assembly of these parts will occur in a proper way. The simple
production of flagellar parts in a random non-specific way will only
produce a junk pile - not a highly complex flagellar system.

Now, of course, if you throw natural selection into the picture, this
is supposed to get evolution out of this mess. It sort through the
potential junk pile options and picks only those assemblages that are
beneficial, in a stepwise manner, until higher and higher systems of
functional complexity are realized. This is how it is supposed to
work. The problem with this notion is that as one climbs up the
ladder of functional complexity, it becomes more and more difficult to
keep adding genetic sequences together in a beneficial way without
having to cross vast gaps of neutral or even detrimental changes.

For example, start with a meaningful English word and then add to or
change that word so that it makes both meaningful and beneficial sense
in a given situation/environment. At first such a game is fairly easy
to do. But, very quickly you get to a point where any more additions
or changes become very difficult without there being significant
changes happening that are "just right". The required changes needed
to maintain beneficial meaning with longer and longer phrases,
sentences, paragraphs, etc., start to really get huge. Each word has
a meaning by itself that may be used in a beneficial manner by many
different types of sentences with completely different meanings.
Although the individual word does have a meaning by itself, its
combination with other words produces an emergent meaning/function
that goes beyond the sum of the individual words. The same thing
happens with genes and proteins. A portion of a protein may in fact
work well in a completely different type of protein, but in the
protein that it currently belongs to, it is part of a completely
different collective emergent function. Its relative order as it
relates to the other parts of this larger whole is what is important.
How is this relative order established if there are many many more
ways in which the relative order of these same parts would not be
beneficial in the least?

Again, just because the right parts happen to be in the same place at
the same time does not mean much outside of a pre-established
information code that tells them how to specifically arrange
themselves.

> > In
> > order to keep up with this exponential decrease in average cone
> > density, the number of blind men has to increase exponentially in
> > order to find the rarer cones at the same rate. Very soon the
> > environment cannot support any more blind men and so they must
> > individually search out exponentially more and more sequence space, on
> > average, before success can be realized (i.e., a cone or cluster of
> > cones is found). For example, it can be visualized as stacked levels
> > of rooms. Each room has its own average density of ice cream cones.
> > The rooms on the lowest level have the highest density of ice cream
> > cones - say one cone every meter or so, on average. Moving up to the
> > next higher room the density decreases so that there is a cone every 2
> > meters or so. Then, in the next higher room, the density decreases to
> > a cone every 4 meters or so, on average. And, it goes from there.
> > After 30 or so steps up to higher levels, the cone density is 1 every
> > billion meters or so, on average.
>
> If the development of each protein started from scratch you may have an
> excellent arguement but nearly all proteins from other proteins so you are
> starting from a point that is known to be functional.

You are actually suggesting here that the system in question had its
origin in many different places. You seem to be suggesting that all
the various parts found as subparts of many different systems somehow
brought themselves together to make a new type of system . . . just
like that. Well now, how did these various different functional
parts, as subparts of many different systems, know how to come
together so nicely to make a completely new system of function? This
would be like various parts from a car simply deciding, by themselves,
to reassemble to make an airplane, or a boat, or a house.

Don't you see, just because the subparts are functional as parts of
different systems of function does not mean that these subparts can
simply make an entirely new collective system of function. This just
doesn't happen although evolutionists try and use this argument all
the time. It just doesn't make sense. It is like throwing a bunch of
words on the ground at random saying, "Well, they all work as parts of
different sentences, so they should work together to make a new
meaningful sentence." Really now, it just doesn't work like this.
You must be able to add the genetic words together in a steppingstone
sequence where each addition makes a beneficial change in the overall
function of the evolving system. If each change does not result in a
beneficial change in function, then nature will not and cannot select
to keep that change. Such non-beneficial changes are either
detrimental or neutral. The crossing of such detrimental/neutral gaps
really starts to slow evolution down, in an exponential fashion,
beyond the lowest levels of specified functional complexity. Very
soon, evolution simply stalls out and cannot make any more
improvements beyond the current level of complexity that it finds
itself, this side of zillions of years of average time.

Sean
www.naturalselection.0catch.com

Bennett Standeven

unread,
Jan 15, 2004, 7:41:15 PM1/15/04
to
seanpi...@naturalselection.0catch.com (Sean Pitman) wrote in message news:<80d0c26f.04011...@posting.google.com>...

> howard hershey <hers...@indiana.edu> wrote in message news:<bu46sv$srt$1...@hood.uits.indiana.edu>...
>
> > > Consider the scenario where there are 10 ice cream cones on the
> > > continental USA. The goal is for a blind man to find as many as he
> > > can in a million years.
> >
> > Except that is NOT what evolution does. Evolution starts with an
> > organism with pre-existing sequences that produce products and interact
> > with environmental chemicals in ways that are useful to the organism's
> > reproduction.
>
> Yes . . . so start the blind man off with an ice-cream cone to begin
> with and then have him find another one.

OK; the ice cream cones are probably found in shops; so given any
cone, odds are that another cone is just a few inches away. This is
still true even if there is only one shop in the USA.


> > Up is good. Down
> > is bad.
>
> Ice-cream cone = Good or "Up" (to one degree or another) or even
> neutral depending upon one's current position as it compares to one's
> previous position. For example, once you have an ice cream, that is
> good. But, all changes that maintain that ice cream but do not gain
> another ice cream are neutral.
>
> No ice-cream cone = "Bad", "Down", or even "neutral" depending upon
> one's current position as it compares to one's previous position.
>
> > Flat is neither good nor bad.
>
> Exactly. Flat is neutral. The more neutral space between each "good"
> upslope/ice-cream cone, the longer the random walk. The average
> distance between each selectable "good" state translates into the
> average time required to find such a selectable state/ice-cream cone.
> More blind men searching, like 10,000 of them, would cover the area
> almost 10,000 times faster than just one blind man searching alone.
> However, at increasing levels of complexity the flat area expands at
> an exponential rate.

But it does not matter, because the blind men always start out in the
ice cream shop, with an ever increasing selection of cones within
arm's reach. Of course, they'll never find any of the other shops, but
so what?

[...]

then each of them is a local high point; only these mesas will have
blind men on them.


> > But you were wondering how something new could arise *after* the blind
> > men are already wandering around the mesas? The answer is that it
> > depends. They can't always do so.
>
> And why not Howard? Why can't they always do so? What would limit
> the blind men from finding new mesas? I mean really, each blind man
> will self-replicate (hermaphrodite blind men) and make 10,000 new
> blind men on the mesa that he/she/it now finds himself on.

But since the current mesa is a local high point, there is nowhere for
them to go.

[...]


> > Let's say that each mesa top has a different basic *flavor* of
> > ice cream. Say that chocolate is a glycoside hydrolase that binds a
> > glucose-based glycoside. Now let's say that the environment changes so
> > that one no longer needs this glucose-based glycoside (the mesa sinks
> > down to the mean level) but now one needs a galactose-based glycoside
> > hydrolase.
>
> You have several problems here with your illustration. First off,
> both of these functions are very similar in type and use very similar
> sequences.

That's not a "problem", it's the whole point. Evolution by definition
involves gradual changes, in which new systems have similar functions
and definitions to the old ones.

> Also, their level of functional complexity is relatively
> low (like the 4 or 5 letter word level).

I don't know exactly what "galactose-based glycoside" is, but
something tells me that it takes more than 4 or 5 amino acids to bind
to it.

> Also, you must consider the likelihood that the environment would change
> so neat so that galactose would come just when glucose is leaving.

Yes. More likely the galactose was always there, but was ignored in
favor of the glucose, until the latter disappeared.

> Certainly if you could program the environment just right, in perfect
> sequence, evolution would be no problem. But you must consider the
> likelihood that the environment will change in just the right way to
> make the next step in an evolutionary sequence beneficial when it
> wasn't before.

That's pretty easy; following the mesa analogy, either the high mesa
drops to be lower than the formerly low one, or the low one rises
above the formerly high one. Happens all the time.

> The odds
> that such changes will happen in just the right way on both the
> molecular level and environmental level get exponentially lower and
> lower with each step up the ladder of functional complexity.

No; the chance that two sequences will interchange in relative fitness
does not depend on how complex they are.

> What was so easy to evolve with functions requiring no more than a few
> hundred fairly specified amino acids at minimum, is much much more
> difficult to do when the level of specified complexity requires just a few
> thousand amino acids at minimum. It's the difference between evolving
> between 3-letter words and evolving between 20-letter phrases. What
> are the odds that one 20-letter phrase/mesa that worked well in one
> situation will sink down with a change in situations to be replaced by
> a new phrase of equal complexity that is actually beneficial? -

Quite good, I'd say. I can easily imagine the relative fitness of
"Today we'll talk about unicorns" exchanging places with that of
"Today we'll talk about unicode", for example. Those are 25-letter
phrases; making them even longer would only increase the number of
nearby phrases with potential value.

> Outside of intelligent design? That is the real question here.
>
> > Notice that the difference in need here is something more
> > like wanting chocolate with almonds than wanting even strawberry, much
> > less jalapeno or anchovy-flavored ice cream. The blind man on the newly
> > sunk mesa must keep walking, of course, but he is not thousands of miles
> > away from the newly risen mesa with chocolate with almonds ice cream on
> > top.
>
> He certainly may be extremely far away from the chocolate with almonds
> as well as every other new type of potentially beneficial ice cream
> depending upon the level of complexity that he happens to be at (i.e.,
> the average density of ice-creams of any type in the sequence space at
> that level of complexity).

Yes; the higher the lever of complexity, the more likely that the new
ice cream cone is nearby, since the fancier (more complex) flavors
tend to appear in the stores with the largest selection.

> > Changing from one glucose-based glycoside hydrolase to one with a
> > slightly different structure is not the same as going from chocolate to
> > jalapeno or fish-flavored ice cream. Not even the same as going from
> > chocolate to coffee. The "island" of chocolate with almonds is *not*
> > going to be way across the ocean from the "island" of chocolate.
>
> Ok, lets say, for arguments sake, that the average density of
> ice-cream cones in a space of 1 million square miles is 1 cone per 100
> square miles. Now, it just so happens that many of the cones are
> clustered together. There is the chocolate cluster with all the
> various types of chocolate cones all fairly close together. Then,
> there is the strawberry cones with all the variations on the
> strawberry theme pretty close together. Then, there is the . . .
> well, you get the point. The question is, does this clustering of
> certain types of ice creams help is the traversing the gap between
> these clustered types of ice creams? No it doesn't. If anything, the
> clustering only makes the average gap between clusters wider. The
> question is, how to get from chocolate to strawberry or any other
> island cluster of ice creams when the average gap is still quite
> significant?

You don't; if you want to get from chocolate to strawberry, you need
to do it early on, when the distance is smaller. That's why
fundamental differences between organisms (such as between chocolate
and strawberry ice cream) are taken as evidence that they are only
distantly related.

>
> You see, the overall average density of cones is still significant to
> the problem no matter how you look at it. Clustering some of them
> together is not going to help you find the other clusters

Who said we had to find all of the clusters?

> > It will be nearby where the blind man is. *And* because chocolate with
> > almonds is now the need, it will also be on the new local high mesa
> > (relative to the position of the blind man on the chocolate mesa). The
> > blind man need only follow the simple rules (Up good. Down bad. Neutral
> > neutral. Keep walking.) and he has a good chance of reach the 'new' local
> > mesa top quite often.
>
> And what about the other clusters? Is the environment going to change
> just right a zillion times in a row so that bridges can be built to
> the other clusters?

No; the blind men at the other clusters reached them when they were
still a part of this one. Eventually the clusters split apart and
"drifted" away from each other. (Much as galaxies "drift" apart due to
cosmic expansion.)

[...]

>
> > My
> > presumption is that the successful search is almost always going to
> > start from the pre-existing mesa
>
> Agreed.
>
> > with the closest flavor to the new need
> > (or from a duplicate, which, as a duplicate, is often superfluous and
> > quickly erodes to ground level in terms of its utility).
>
> This is where we differ. Say you have chocolate and vanilla. Getting
> to the different varieties of chocolate and vanilla is not going to be
> much of a problem. But, say that neither chocolate nor vanilla are
> very close to strawberry or to each other. Each cluster is separated
> from the other clusters by thousands of miles. Now, even though you
> already have two clusters in your population, how are you going to
> evolve the strawberry cluster if an environmental need arises where it
> would be beneficial?

In that case, you wouldn't. You'd have to settle for chocolate ice
cream with strawberries or some such.

Similarly, we would not expect birds to evolve jet engines, as they
are too different from any system the birds possess now.

[...]


> The suggestion that absolutely all of the clusters are themselves
> clustered together in a larger cluster or archipelago of clusters in a
> tiny part of sequence space is simply a ludicrous notion to me -
> outside of intelligent design that is. Oh no, you, Robin, Deaddog,
> Sweetness, Musgrave, and all the rest will have to do a much better
> job and explaining how all the clusters can get clustered together
> (when they obviously aren't) outside of intelligent design.

It isn't necessary that they _all_ be clustered in that fashion; only
that some of them are.

Bennett Standeven

unread,
Jan 15, 2004, 7:46:23 PM1/15/04
to
dfo...@gl.umbc.edu (david ford) wrote in message news:<b1c67abe.04011...@posting.google.com>...


> What I meant to say was, there's no such thing as an intelligent
> designer of computer programs.

Heh. Sometimes it feels that way, even with my own programs.

Chris Merli

unread,
Jan 15, 2004, 9:59:07 PM1/15/04
to

Actually this comes from the examination of many protein sequences.

So how would you explain that there are hundreds of examples of parts of
proteins that were obviously lifted from other proteins. More importantly
how do you explain the nested heiarchies that they follow. A designer may
borrow an idea to use again but they would not modify simple highly
functional components for each new use. That would be like completely
re-engineering a bolt for every new machine. So then your theory must
predict that we would find very similiar proteins in very diverse organisms.
Is this a prediction of your theory?

I thought you were beyond this base level of a strawman arguement.

>
> Don't you see, just because the subparts are functional as parts of
> different systems of function does not mean that these subparts can
> simply make an entirely new collective system of function. This just
> doesn't happen although evolutionists try and use this argument all
> the time. It just doesn't make sense.

Actually if you look at the components of the clotting system or the globin
genes you will see that this is exactly what happens. And if you want to go
for real word scrabble try the immune system. The basic idea is to shuffle
the parts of these genes to creat hundreds of different antibodies.

It is like throwing a bunch of
> words on the ground at random saying, "Well, they all work as parts of
> different sentences, so they should work together to make a new
> meaningful sentence." Really now, it just doesn't work like this.

Maybe it would help if you avoided using analogies and just stuck to
biological examples.

> You must be able to add the genetic words together in a steppingstone
> sequence where each addition makes a beneficial change in the overall
> function of the evolving system. If each change does not result in a
> beneficial change in function, then nature will not and cannot select
> to keep that change. Such non-beneficial changes are either
> detrimental or neutral. The crossing of such detrimental/neutral gaps
> really starts to slow evolution down, in an exponential fashion,
> beyond the lowest levels of specified functional complexity. Very
> soon, evolution simply stalls out and cannot make any more
> improvements beyond the current level of complexity that it finds

> itself, this side of zillions of years of average time.
>
> Sean
> www.naturalselection.0catch.com

I noticed that you did not address the most important parts of my last post.
If you have developed a scientific theory as an alternative explanation then
you should be able to provide some testable predictions to support the
theory.

>


Von Smith

unread,
Jan 16, 2004, 12:33:18 AM1/16/04
to
seanpi...@naturalselection.0catch.com (Sean Pitman) wrote in message news:<80d0c26f.04011...@posting.google.com>...

Except of course that organisms that actually have all these
components don't actually produce a junkpile, and in many cases, such
as in the TTSS or Tsp pilus, the relevant parts already assemble in
substantially the same way they do for a flagellum. It appears that
Dr. Pitman has taken his strawman version of evolution to the next
level: not content with suggesting that proteins must evolve from
scratch from random peptide sequences, he is now telling us that
complex multi-protein systems must evolve from random junk-piles of
constituent parts.

I would have been more impressed if you had written this *after*
giving a substantive reply to Deaddog's recent excellent post on
Synthetic Biology, which probably sheds some light on how biologists
*really* think complex multi-protein systems might evolve. In it, he
cites a paper in which researchers randomly switched around some of
the parts involved in complex multi-protein interactions to see what
they would do.

Combinatorial synthesis of genetic networks.
Guet CC, Elowitz MB, Hsing W, Leibler S.
Science. 2002 May 24; 296(5572): 1466-70.

http://www.sciencemag.org/cgi/content/full/296/5572/1466

So what happens when one shakes up the regulatory bits of a biological
system and lets them fall where they will? AIUI, far from ending up
with nothing but random junkpiles, the researchers were able to obtain
a variety of novel logically-functioning phenotypes. No need for some
pre-existing homonculus magically prompting the various parts on how
to behave: as often as not the parts were able to associate and
interact coherently left to their own devices. Of course it is
possible that this liberal arts major is misunderstanding the article.
Perhaps the biologically washed can comment more coherently.

>
> Now, of course, if you throw natural selection into the picture, this
> is supposed to get evolution out of this mess. It sort through the
> potential junk pile options and picks only those assemblages that are
> beneficial, in a stepwise manner, until higher and higher systems of
> functional complexity are realized. This is how it is supposed to
> work. The problem with this notion is that as one climbs up the
> ladder of functional complexity, it becomes more and more difficult to
> keep adding genetic sequences together in a beneficial way without
> having to cross vast gaps of neutral or even detrimental changes.

Maybe, maybe not. Again, you wouldn't necessarily need an
astronomical amount of novel assembly to get motility out of a Tsp
pilus; many of the constituent parts are not only already there, but
are already interacting in the way they do in a flagellum. It appears
that once again, you are assuming that we are talking about evolving
such a system from *any* random assemblage of the individual
components, rather than from a logical precursor like a pilus. And
besides, this recent work in systems biology indicates that even if we
*are* talking about randomly rejumbling the components of a system,
the prospects of getting a novel beneficial function as a result may
not be quite as grim as you make out.

<snip yet another English language analogy>

I may be somewhat out of my depth here, but what the hell: Proteins
interact with one another according to chemistry. Since the precursor
proteins had basically the same chemical propoerties they do now, they
already more or less "knew how" to interact with one another. You
might want a point mutation or two to improve affinity; this is hardly
a problem for evolution. Timing and delivery of the parts is
controlled by things like regulatory sequences and transport proteins;
these can also evolve new behaviors, and have been observed to do so.
Genes can be up- or down-regulated, and regulatory sequences can
evolve to respond to different repressors. Regulatory sequences can
be switched around. Transport proteins can be co-opted and modified to
transport different substances.

ISTM you are trying to create a mystification. We don't have all the
*specific* answers to how, exactly, this or that structure evolved, or
even know all the details about how the various parts of the flagellum
work, but I don't think that how proteins generally "know how" to
interact with one another is the sort of inexplicable black magic you


seem to think it is.

Von Smith
Fortuna nimis dat multis, satis nulli.

howard hershey

unread,
Jan 16, 2004, 2:13:32 PM1/16/04
to

Sean Pitman wrote:

> howard hershey <hers...@indiana.edu> wrote in message
> news:<bu46sv$srt$1...@hood.uits.indiana.edu>...
>
>
>>> Consider the scenario where there are 10 ice cream cones on the
>>> continental USA. The goal is for a blind man to find as many as
>>> he can in a million years.
>>
>> Except that is NOT what evolution does. Evolution starts with an
>> organism with pre-existing sequences that produce products and
>> interact with environmental chemicals in ways that are useful to
>> the organism's reproduction.
>
>
> Yes . . . so start the blind man off with an ice-cream cone to begin
> with and then have him find another one.
>
>
>> The situation is more like 10,000 blind men in a varying topography
>> who blindly follow simple and dumb rules of the game to find useful
>> things (ice cream at the tops of fitness peaks):
>
>
> You don't understand. In this scenario, the positively selectable
> topography is the ice-cream cone.

The reason why I used a mesa loaded with ice cream cones is because of
the difference in size of the searching blind man (the modal sequence in
a specific population) and an ice cream cone (all sequences with the
same effective functional activity).

The only way your scenario would be an accurate reflection of reality
is if the ice cream cone were really an ice cream mountain that the
blind woman can climb, with a few dribs and drabs of ice cream (of some
flavor) at the base, with increasing concentrations of ice cream up the
slope toward the mesa, enticing the man upward to the mesa she can
wander around.

Now, why do I use this model rather than your sudden tiny ice cream
cones that pop out of nowhere as tiny dots on the tops of telephone
poles in monotonously flat landscape? You need to remember what these
entities are representing and what the reality of a search through
sequence space in real protein sequence space would look like.

The blind man following my dumb rules is the sequence du jour (the
current modal sequence) of a population of organisms. The reward for
following the rules is winding up on the mesas where the maximum utility
or reward to the organism (measured by the metric of reproductive
success; ice cream is just like sex) is. This is a mesa rather than an
alp because there are literally thousands of sequences that can have
effectively the same optimal functionality, as evidenced by the fact
that there are hundreds to thousands of different sequences that perform
the same function with effectively equal efficiency in different species
of modern organisms and even within species. That doesn't preclude
minor variations in altitude on the mesa top. Moreover, it is quite
clear that there are also many sequences that have *less* utility than
the optimal utility at the mesa. The mesa is surrounded by ground that
*slopes* upward toward the mesa.
That is, the fitness mesa does not simply pop up straight out of
ground (it is unlike most mesas in this sense) like Devil's Tower (WY)
but there are many sequences of varying utility from no selectable
utility through partial utility to optimal utility. It is also clear
that optimal utility is a relative condition because many enzymes and
systems have to balance conflicting needs (such as need for being able
to utilize several different substrates).


> There are no other selectable fitness peaks here. The rest of the
> landscape is neutral.

For a *given* function or utility, the vast majority of the landscape
will be neutral (meaning equally useless in this case) for *that*
function. However, nearby the functional mesa will be other mesas that
have *related* functions or utility. These may even be poking out from
the gradual slope leading up to the function or utility of current
interest. For example, nearby a hypothetical beta galactosidase mesa
will likely be mesas that bind other sugar-adducts via a glycoside
linkage and hydrolyse those linkages, but do not bind galactose. That
is, there will be sequences which already are part way up the slope
leading to the lactase activity, but the mesas part ways and go upward
in a different directions (one toward glucose-adducts, say). The
reason these sequences are *clustered* close to the sequences for
galactosidase activity are *because* cleavage of a galactose-adduct bond
shares many of the structural and sequence feature needs with activities
that cleave glucose-adduct bonds. These sequences are clustered because
they are similar, like chocolate and chocolate with nuts.

Now this is a very different type of sequence landscape than the one I
see Sean proposing. Let me try to ascii draw what I see as the
differences between Sean's model of sequence space and mine. I could be
wrong about his model since he keeps using word descriptions that
disagree with his mathematical model, which invariably assumes that what
determines the difficulty of evolving a new function is the distance
between some *average* or *random* sequence and the new sequence that
must be generated, a point he repeatedly makes but denies making.

Howard's interpretation of Sean's model of sequence space:

|
|
|
| . .
|
| x
|
|
|
| . .
|
|
| .
|
|
| .
|
| o
|
|
| .
|
|
|
|
| . .
|
|
|_______________________________________________________

By 'sequence space' I specifically mean *all* possible protein sequences
of a particular length, not just those with, say, lactase activity.
The .'s in this model represents the 'ice cream cones'; that is, the
rare sequences that serve *any* useful function whatsoever. Everywhere
else we have a flat surface. The x represents the function (the type of
ice cream cone) you think the blind man (the o) must find by wandering
around the flat spaces. The blind man (the o) is the modal sequence or
starting sequence in the search. I started the o at a random or average
site in all sequence space because that is what Sean's *mathematical*
treatment presumes. He presumes that the search from an average or
random sequence to the desired sequence, which on average would depend
on the overall ratio of useless or neutral sequence to useful sequence
is what is important. Moreover, the ratio Sean uses in his calculations
is the ratio of sequences for a *particular* useful function to *all*
sequence space, and thus any sequence which is useful for a different
function other than the chosen one is put in the denominator as being
equivalent to a sequence that has no utility whatsoever. That is one
the reason why I consider his goal to be a teleological or
pre-determined role.

What I cannot represent here, but is certainly an important point, is
the idea that the .'s in Sean's model are completely randomly
distributed wrt to function. That is, if the . at the lower right
encodes a glucose-based glycoside hydrolase, a . representing a
galactose-based glycoside hydrolase will NOT cluster with the
glucose-based glycoside hydrolase, but will be found at some random
position (on average, far away) in this sequence space wrt the sequence
that encodes glucose-based glycoside hydrolase. In this model, and only
in this model where the search involves a completely random search, the
separation between functional sequences is a function of the ratio of
useful to non-useful sequences and nothing else. Neither the starting
point of the blind man nor nor any of the useful sequences show any
clustering of functionalities. Feel free, Sean, to correct any part of
this model that you regard as a misrepresentation of what your
*mathematical* model presents.


Howard's model of sequence space.

|
| _______ _______
| / .o. \ / ... \
| | ... | | .o. _|_
| \ ... / \ .../xxx\
| ------- -----|xxx|
| \xxx/
| _______ _______
| / ... \ / ... \
| | .o. | | ... |
| \ ... / \ .o. /
| -------___ -------
| / ... \
| | ... |
| \ o.. / _______
| ------- / ... \
| | ... |
| \ ... /
| -------
| _______
| / o.. \
| | ... |
| \ ... /
| -------
| _______ _______
| / ... \ / ... \
| | ..o | | .o. |
| \ ... / \ ... /
| ------- -------
|_______________________________________________________

By 'sequence space' I specifically mean *all* possible protein sequences
of a particular length. The .'s represents the 'ice cream cones'; that
is, the rare sequences that serve *any* useful function whatsoever.
There are a number of sequences that are equally useful. I have
clustered these in a 3x3 dot array, because the equally useful sequences
are close together. Of course, the reality would be that the size of
these boxes will be highly variable. Some will be only a very small
cluster. Others, like fibrinogen peptide, can essentially cover the
entire sequence space! There is little relationship between size of the
protein and the number of sequences in sequence space that can fulfill
that function. But, in general, the smaller proteins tend to have
higher constraint (fewer sequences will fullfil the function). The
differences between the .'s are selectively neutral, so the position of
the blind man (the modal sequence in a particular organism) is random
within that group. There is *also* a pneumbra of surrounding sequences
with *less* utility of varying degrees of full functionality. I have
represented that by the box around the cluster of useful sequences. The
boundary is the edge of selectable utility (where the blind man starts
to notice a selective slope). The x represents the function (the type of
ice cream cone) you think the blind man (one of the o's) must find.
Notice that this representation is a representation of a real landscape
with real topography, not a perfectly flat plane with telephone poles
sticking up randomly at scattered points.

I am starting with a real cell and not with a hypothetical blind man
who is starting as some random sequence in all of sequence space.
Each of my blind men (genes, if you will) already occupies a site on the
mesa of functionality, but the different mesas (and their modal gene
sequences) represent quite different functionalities. One peak may
represent a glucose-glycoside hydrolase (the one on the upper right very
near the xxx mesa). Another, down on the lower left, may represent a
sequence with fatty acid synthetase activity. The whole board
represents all of sequence space, after all. But the o's of my modal
gene sequences in populations are not on random or average positions.
They are specifically on mesas of functionality. I would argue that
that is a better representation of reality than Sean's (or the best I
can make of Sean's) model of reality.

Notice that there is also some overlap in functionalities. And there is
even one potentially useful site that has no blind man (middle far
right). This is, simply put, a potential function that this particular
cell does not have, but does exist in sequence space, such as nylonase
activity or the ability to extract energy from H2S. It is certain that
this cell does not *need* this activity for survival. It is not
necessarily the case that it could not *use* it, although that may also
be true. One does not, after all, *need* nylonase activity in all
possible environments. In my model, all the blind men are moving around
their respective mesas. Some may even take a few steps downslope by
accident. It is highly unlikely that a *randomly chosen* one of these
10,000 pre-existing blind men (and 10,000 does not seem to be an absurd
number for the number of different genes) will find, by such a walk, the
spots marked xxx.

But evolution does not work by a *randomly* chosen or *average* blind
man wandering through functionless space to chance upon the xxx's in my
model of sequence space. In particular, in my model, there is, compared
to Sean's model, a definite, obvious, and intuitive clustering of
functionally useful sequences. That is, the cluster that overlaps the
xxx sequence is not some random sequence with some random function. It
is, let's say, a glucose-based glycoside hydrolase with no selectable
beta galactosidase activity. In my model, such a hydrolase is not
randomly present in the sequence space, but is specifically likely to be
clustered close to those sequences that do have selectable beta
galactosidase activity.

In fact, even if one started with a randomly chosen blind man starting
at some place on the flats between useful sequences, if that sequence
were ever to become useful, it would do so by climbing the nearest mesa
that has no blind man on top (that blind man's landscape does not
include already occupied mesas) using the simple rules I described.

Another thing that is not represented diagramatically is the role of
duplication. A duplicate of the blind man on the mesa close to the
xxx's does not have the same position as the original blind man (which
is already at the top of the mesa). It is, instead, often at a position
that is close to the flatland (that is, one copy, the one I call the
'duplicate' is functionally redundant rather than functionally useful).
Thus, when this redundant blind man takes a step toward the xxx's he
is not taking a step down, but a step up. The landscape for this man is
different than the landscape for the identical clone of this man. This,
of course, is hard to represent in a simple plane.

> Some of the ice-cream cones may be more
> positively selectable than others (i.e., perhaps the man likes
> vanilla more than chocolate). However, all positive peaks are
> represented in this case by an ice-cream cone.
>
>
>> Up is good. Down is bad.
>
>
> Ice-cream cone = Good or "Up" (to one degree or another) or even
> neutral depending upon one's current position as it compares to one's
> previous position. For example, once you have an ice cream, that is
> good. But, all changes that maintain that ice cream but do not gain
> another ice cream are neutral.
>
> No ice-cream cone = "Bad", "Down", or even "neutral" depending upon
> one's current position as it compares to one's previous position.
>
>
>> Flat is neither good nor bad.

This position appears to represent ice cream cones as an all-or-nothing
phenomenon. There are no possible intermediate states in this model.
It looks like a flat plain with telephone poles. In short, it looks
like an artificial landscape, not a real one.


>
> Exactly. Flat is neutral. The more neutral space between each
> "good" upslope/ice-cream cone, the longer the random walk. The
> average distance between each selectable "good" state translates into
> the average time required to find such a selectable state/ice-cream
> cone. More blind men searching, like 10,000 of them, would cover the
> area almost 10,000 times faster than just one blind man searching
> alone. However, at increasing levels of complexity the flat area
> expands at an exponential rate.

How does one determine, in mathematical terms, "level of complexity"?
The reason why I did not have landscapes where I used sequence space at
a given level of complexity rather than at a given amino acid number is
that I have no idea how one determines "level of complexity".
Why do I have to keep asking that question? And what is your evidence
that increasing levels of complexity causes a change in the ratio of
utile to useless sequence? How do you determine the ratio of utile (for
*any function*) to useless (for *any function*) sequence in any case?
What is it that prevents clustering of functionally related sequences in
your landscape?

> In order to keep up and find new
> functions at these higher levels of functional complexity, the
> population of blind men will have to increase at an equivalent rate.

Only if you think the blind men start at random positions and go to a
sequence which is randomly placed wrt their position.

> The only problem with increasing the population is that very soon the
> local environment will not be able to support any larger of a
> population. So, if the environment limits the number of blind men
> possible to 10,000 - that's great if the average neutral distance
> between ice-cream cones in a few miles or so, but what happens when,
> with a few steps up the ladder of functional complexity, the neutral
> distance expands to a few trillion miles between each cone, on
> average? Now each one of your 10,000 blind men have to search around
> 50 million sq. miles, on average, before the next ice-cream cone or a
> new cluster of ice cream cones will be found by even one blind man in
> this population.

Could you explain the relevance of the above model to the real world?
Why does the *average* distance between a *random* site and a
*teleologically determined* site matter? Wouldn't the distance between
the blind man closest to a teleologically determined site and that site
be more important and relevant? We are not interested in the odds of
the *average* sequence changing into the teleologically determined one.
We are interested in the *best* odds of *any* existing sequence
changing into the teleologically determined one. The best odds are
those of the pre-existing sequence that is closest to the end sequence
and has nothing to do with the odds of an average or random sequence
becoming the end sequence.

>> Keep walking in all cases.
>
>
> They keep walking alright - a very long ways indeed before they reach
> anything beneficially selectable at anything very far beyond the
> lowest levels of functional complexity.

Only if they started from random spots in sequence space. If the blind
man which is closest to the xxx starts walking, it will quickly, by the
simple rules I invoked, find its way up the Mt. Improbable right next door.

>> It would not take too long for these 10,000 blind men to be found
>> in decidedly non-random places (the high mesas of functional
>> utility where they are wandering around the flat tops if you
>> haven't guessed).


> There is a funny thing about these mesas. At low levels of
> complexity, these mesas are not very large. In fact, many of them
> are downright tiny - just one or two steps wide in any direction and
> a new, higher mesa can be reached. However, once a blind man finds
> this new mesa new higher mesa (representing a different type of
> function at higher level of specified complexity) and climbs up onto
> its higher surface, the distance to a new mesa at the same height or
> taller is exponentially greater than it was at the lower levels of
> mesas.
>
> ___ __ __ _
> _-_ __ __-_ _-_- -__-__-_- _-__-_-_-__-
> -_- _-_-_ _-_-__
>

Can you make the above make sense? Remember in my model, that the usual
way for a blind man to move involves a change in the landscape or the
presence of a duplicate blind man who is now redundant and for whom the
landscape is differently shaped.

>> And the ice cream cones (the useful functions), remember, are not
>> randomly distributed either. They are specifically at the tops of
>> these mesas as well. That is what a fitness landscape looks like.
>
>
> Actually, the mesa itself, every part of its surface, represents an
> ice cream cone. There is no gradual increase here. Either you have
> the ice-cream cone or you don't.

I.e., your model is of a flat plain with telephone poles where there
cannot be intermediacy in function. Where an enzyme cannot mutate or
generate a closely related sequence with 50% of optimal activity. Or
10%. In your model, it is indeed all-or-nothing. That is what I get
from this discussion. Am I right? [The reason I ask is because I will
want to compare your and my model with reality -- that is test the model
against the evidence of nature -- to see which is closer to the way that
real organisms and real enzymes and real systems of change work.]

> If you don't have one that is even
> slightly "good"/beneficial, then you are not higher than you were to
> begin with and you must continue your random walk on top of the flat
> mesa that you first started on (i.e., your initial beneficial
> function(s)).

"Good/beneficial" is not an absolute value. It is a relative value. It
is "better than". Indeed, in an *unchanged* selective environment, it
is unlikely that there will be a mesa of higher utility arising out of
an original mesa that will not have already been discovered by a random
walk which retains the original activity or function at each step. What
your model seems to indicate is something quite different. Your
landscape is like a flat plain with telephone poles, and you seem to say
that the only way to reach a new telephone pole is to climb down and
wander the flatlands blindly. That is, one first completely loses all
functional utility and wanders functionless space.

In my model, it may be that there is, in fact, a mesa newly arisen out
of an original mesa that suddenly looks more attractive than the
original. This would be a consequence of a change in environment. An
example of this would be the conversion of ebg to lactase due to a
change in environment that made lactase activity far more beneficial
than the original activity of ebg. Or it could be due to the production
of a redundant duplicate, with the duplicate being free to explore new
nearby upward directions that the original could not, because its
function was too valuable. Or it could be a new function for an old
protein that appears by a change in regulation (as in eye crystallins).
But, then, I don't see any changes that must necessarily involve long
selectively neutral walks. I only see walks to related structures in a
cluster that has related functions or emergent functions of old
structures. And I envision a *real* landscape, not a flat plain with
telephone poles.

>> If this topography of utility only changed slowly, at any given
>> time it would appear utterly amazing to Sean that the blind men
>> will all be found at these local high points or optimal states (the
>> mesas licking the ice cream cones on them) rather than being
>> randomly scattered around the entire surface.
>
>
>
> If all the 10,000 blind men started at the same place, on the same
> point of the same mesa, and then went out blindly trying to find a
> higher mesa than the one they started on, the number that they found
> would be directly proportional to the average distance between these
> taller mesas.

My model does no such thing. The 10,000 blind men are found on
functionally useful mesas that are scattered *in clusters* throughout
sequence space. You seem to think that I am thinking that each blind
man represents an individual organism. I am thinking of each blind man
representing a modal sequence in a population of organisms and their
scatter representing the pattern of real functional cell sequences in
sequence space. That is because evolution of new function by sequence
change does not start from some arbitrary set of random sequences. It
starts with already useful sequences in already functioning cells. Each
mesa does something different; each cluster in a mesa or cluster of
related mesas does something related to what other members of the
cluster do. Each blind man is moving around his functional mesa.
Probability says that the blind man in the cluster closest to the new
sequence is the most likely one to find a new mesa optimum. Not some
random blind man. The average density of mesas is irrelevant to the
odds of some blind man finding a new solution or sequence. All that
matters is how far the nearest sequence with a blind man is and whether
the environmental landscape has changed to favor movement from current
optimi.

> If the density of taller mesas, as compared to the one

> they are now on, happens to be, say, one every 100 meters, then they

> will indeed find a great many of these in short order. However, if
> the average density of taller mesas, happens to be one every 10,000
> kilometers, then it would take a lot longer time to find the same
> number of different mesas as compared to the number the blind men
> found the first time when the mesas were just 100 meters apart.

My whole point is that *average* distance from an *average* blind man is
utterly irrelevant to reality. It is not wrong. It is irrelevant.

>> They reached these high points (with the ice cream) by following a
>> simple dumb algorithm.
>
>
> Yes - and this mindless "dumb" algorithm works just fine to find new
> and higher mesas if and only there is a large average density of
> mesas per given unit of area (i.e., sequence space). That is why it
> is easy to evolve between 3-letter sequences. The ratio/density of
> such sequences is as high as 1 in 15. Any one mutating sequence will
> find a new 3-letter sequence within 15 random walk steps on average.
> A population of 10,000 such sequences (blind men) would find most if
> not all the beneficial 3-letter words (ice-cream cones) in 3-letter
> sequence space in less than 30 generations (given that there was one
> step each, on average, per generation).

Notice that you are starting with a *random* 3-letter sequence and
asking how many steps would be required for it to reach another *random*
specified 3-letter sequence by a *random* walk with no intermediate
utility. That is the mathematical argument you repeatedly say you are
NOT making, but repeatedly insist on doing. That model is not wrong.
It is irrelevant.

> This looks good so far now doesn't it? However, the problems come as
> you move up the ladder of specified complexity. Using language as
> an illustration again, it is not so easy to evolve new beneficial
> sequences that require say, 20 fairly specified letters, to transmit
> an idea/function. Now, each member of our 10,000 blind men is going
> to have to take over a trillion steps before success (the finding of
> a new type of beneficial state/ice cream cone) is realized for just
> one of them at this level of complexity.

This does fit the model I presented as my interpretation of what you
said. I think that model has very little relationship to either the
reality of sequence space or the mechanisms of evolution. It is nothing
but the tornado whipping together a 747 argument gussied up so she
doesn't look like the old decrepit whore she is.


>
> Are we starting to see the problem here? Of course, you say that
> knowledge about the average density of beneficial sequences is
> irrelevant to the problem, but it is not irrelevant unless you, like
> Robin, want to believe that all the various ice-cream cones
> spontaneously cluster themselves into one tiny corner of the
> potential sequence space AND that this corner of sequence space just
> so happens to be the same corner that your blind men just happen to
> be standing in when they start their search. What an amazing stroke
> of luck that would be now wouldn't it?

I do think that sequences cluster by functional attributes. That is,
enzymes that hydrolyze glycoside linkages will all have similar
sequences or at least sequences that produce similar 3-D structures with
a *few* key sites being strongly conserved. Why do you think otherwise?

>> But you were wondering how something new could arise *after* the
>> blind men are already wandering around the mesas? The answer is
>> that it depends. They can't always do so.
>
>
> And why not Howard? Why can't they always do so? What would limit
> the blind men from finding new mesas?

The fact that the blind men (the modal sequence of a population) are
already on mesas of utility. Usually a change in functional or
selective landscape is required in the vicinity of a blind man to allow
him to reach a different peak by following the simple rules.

> I mean really, each blind man
> will self-replicate (hermaphrodite blind men) and make 10,000 new
> blind men on the mesa that he/she/it now finds himself on. This new
> population would surely be able to find new mesas in short order if
> things worked as you suggest.

If the change is positively *selective*, the walk of the blind man (the
modal population sequence) to the goal will indeed be rapid. But
neutral drift of a modal population sequence is not a fast process. If
it requires a few steps downward before hitting a new upward slope to a
different function the process will be quite episodic.

> But the problem is that if the mesas
> are not as close together, on average, as they were at the lower
> level where the blind men first started their search, it is going to
> take longer time to find new mesas at the same level or higher. That
> is the only reason why these blind men "can't always" find "something
> new". It has to do with the average density of mesas at that level.

Average density of a specified end only has meaning if one is
envisioning evolutionary searches as a random search for a specified end
from a random or average position. Evolutionary searches that succeed
never or rarely (nylonase, perhaps) start from a random or average site.
They start from a site close to the destination. And since functional
sequences do seem to be clustered rather than randomly scattered across
sequence space, it is not unusual for the starting point of *successful*
evolutionary inventions to be nearby.

>> But remember that these pre-existing mesas are not random places.
>> They do something specific with local utility.
>
>
> The mesas represent sequences with specific utilities. These
> sequences may in fact be widely separated mesas even if they happen
> to do something very similar. Really, the there is no reason for the
> mesas to be clustered in one corner of sequence space. A much more
> likely scenario is for them to be more evenly distributed throughout
> the potential sequence space.

Chose your poison. If mesas are clustered, in fact, reaching a new mesa
that is far away from any cluster becomes more difficult because o's
(the blind men or modal population sequences) are clustered on
pre-existing mesas. Perhaps requiring a chance event like the one that
produced nylonase or one that caused the formation of a chimeric protein
rather than a stepwise change of single nucleotides, as would be
possible if the new mesa were in the same functional family. Or perhaps
making that change impossible for that organism.

If mesas are *evenly* spread throughout sequence space, that still
doesn't obviate the fact that the distance between the *average*
pre-existing mesa, with its blind man, and the new mesa is irrelevant
compared to the distance between the *nearest* pre-existing mesa and the
new mesa. Evolution to the new mesa won't come from some *average* mesa
or some mesa on the other side of sequence space. It will come from
mesas that are closest to the new one.

> Certainly there may be clusters of
> mesas here and there, but on average, there will still be a wide
> distribution of mesas and clusters of mesas throughout sequence space
> at any given level. And, regardless of if the mesas are more
> clustered or less clustered, the *average* distance between what is
> currently available and the next higher mesa will not be
> significantly affected.

No. It will be utterly irrelevant.

>> Let's say that each mesa top has a different basic *flavor* of ice
>> cream. Say that chocolate is a glycoside hydrolase that binds a
>> glucose-based glycoside. Now let's say that the environment
>> changes so that one no longer needs this glucose-based glycoside
>> (the mesa sinks down to the mean level) but now one needs a
>> galactose-based glycoside hydrolase.
>
>
> You have several problems here with your illustration. First off,
> both of these functions are very similar in type and use very similar
> sequences.

No kidding! Who would have thunk that blind evolution would chose to
evolve a lactase from a closely related sequence rather than from some
random or average sequence or from an alcohol dehydrogenase? Surely not
Sean.

> Also, their level of functional complexity is relatively
> low (like the 4 or 5 letter word level). Also, you must consider
> the likelihood that the environment would change so neat so that
> galactose would come just when glucose is leaving. Certainly if you
> could program the environment just right, in perfect sequence,
> evolution would be no problem.

A concentration gradient would suffice; that would provide environments
in which the original strain could grow and also a new niche would be
open for any variant able to exploit it. The environment, of course,
only selects among existing variants, so the selectable change would
have to have already happened.

> But you must consider the likelihood
> that the environment will change in just the right way to make the
> next step in an evolutionary sequence beneficial when it wasn't
> before. The odds that such changes will happen in just the right way
> on both the molecular level and environmental level get exponentially
> lower and lower with each step up the ladder of functional
> complexity.

How does one calculate "functional complexity" so that one knows what
rung of the ladder one is talking about?

> What was so easy to evolve with functions requiring no
> more than a few hundred fairly specified amino acids at minimum, is
> much much more difficult to do when the level of specified complexity
> requires just a few thousand amino acids at minimum.

What do these numbers of amino acids mean wrt "level of specified
complexity". How does one determine that there are a "few hundred
fairly specified amino acids" required for a change in function.
Especially since function can change without changing *any* amino acids
(see the eye crystallins).

> It's the
> difference between evolving between 3-letter words and evolving
> between 20-letter phrases. What are the odds that one 20-letter
> phrase/mesa that worked well in one situation will sink down with a
> change in situations to be replaced by a new phrase of equal
> complexity that is actually beneficial? - Outside of intelligent
> design? That is the real question here.

Well, it would help if you would actually tell us what your meaningless,
gobbledygook, hand-waving terms actually meant and how they could be
operationally quantified.

>> Notice that the difference in need here is something more like
>> wanting chocolate with almonds than wanting even strawberry, much
>> less jalapeno or anchovy-flavored ice cream. The blind man on the
>> newly sunk mesa must keep walking, of course, but he is not
>> thousands of miles away from the newly risen mesa with chocolate
>> with almonds ice cream on top.
>
>
> He certainly may be extremely far away from the chocolate with
> almonds as well as every other new type of potentially beneficial ice
> cream depending upon the level of complexity that he happens to be at
> (i.e., the average density of ice-creams of any type in the sequence
> space at that level of complexity).

That is certainly counter-intuitive and counter-evidence that related
functions tend to be found to have related sequences (be in gene families).

>> Changing from one glucose-based glycoside hydrolase to one with a
>> slightly different structure is not the same as going from
>> chocolate to jalapeno or fish-flavored ice cream. Not even the same
>> as going from chocolate to coffee. The "island" of chocolate with
>> almonds is *not* going to be way across the ocean from the "island"
>> of chocolate.
>
>
> Ok, lets say, for arguments sake, that the average density of
> ice-cream cones in a space of 1 million square miles is 1 cone per
> 100 square miles. Now, it just so happens that many of the cones are
> clustered together. There is the chocolate cluster with all the
> various types of chocolate cones all fairly close together. Then,
> there is the strawberry cones with all the variations on the
> strawberry theme pretty close together. Then, there is the . . .
> well, you get the point. The question is, does this clustering of
> certain types of ice creams help is the traversing the gap between
> these clustered types of ice creams?

It certainly reduces the distance needed to go from chocolate to
chocolate with almonds. But why would anyone think that evolution works
by converting an alcohol dehydrogenase into a glycoside hydrolase rather
than by modifing one glycoside hydrolase into a different one?

> No it doesn't. If anything,
> the clustering only makes the average gap between clusters wider.
> The question is, how to get from chocolate to strawberry or any other
> island cluster of ice creams when the average gap is still quite
> significant?

What evidence do you have that one ever *needs* to convert vanilla into
chocolate? There are, of course, evolutionary mechanisms for making
vanilla/chocolate swirl (chimeric hybrid formation). But we who prefer
our hypotheses to be realistic leave the converting of vanilla into
chocolate to the alchemists and magicians.

> You see, the overall average density of cones is still significant to
> the problem no matter how you look at it. Clustering some of them
> together is not going to help you find the other clusters - unless
> absolutely all of the ice cream islands are clustered together as
> well in a cluster of clusters all in one tiny portion of the overall
> potential space. This is what Robin is trying to propose, but I'm
> sorry, this is an absolutely insane argument outside of intelligent
> design. How is this clustering of clusters explained via mindless
> processes alone?

No. I strongly suspect he is proposing a situation close to that I am
proposing. Where cells have functions and these functions are clustered
and sequences form gene families rather than being scattered in sequence
space. And where the relevant distance is between some current sequence
closest to the end sequence rather than the distance between some random
current sequence and the end sequence.


>
>
>> It will be nearby where the blind man is. *And* because chocolate
>> with almonds is now the need, it will also be on the new local high
>> mesa (relative to the position of the blind man on the chocolate
>> mesa). The blind man need only follow the simple rules (Up good.
>> Down bad. Neutral neutral. Keep walking.) and he has a good chance
>> of reach the 'new' local mesa top quite often.
>
>
> And what about the other clusters? Is the environment going to
> change just right a zillion times in a row so that bridges can be
> built to the other clusters?

You obviously have missed the fact that I am modelling a real organism
which has more than one gene sequence. Who knows what you are modelling?


>
>
>> And remember that there is not just one blind man on one mesa in
>> this ocean of possible sequences. There are 10,000 already present
>> on 10,000 different local mesas with even more flavors than the 31
>> that most ice cream stores offer. Your math always presuposes that
>> whenever you need to find, say, vanilla with cherry the one blind
>> man starts in some random site and walks in a completely random
>> fashion (rather than by the rules I pointed out) across half the
>> universe of sequence space to reach your pre-determined goal by
>> pure dumb luck to find the perfect lick.
>
>
> That is not my position at all as I have pointed out to you numerous
> times. It seems that no matter how often I correct you on this straw
> man caricature of my position you make the same straw man
> assertions. Oh well, here it goes again.
>
> I'm perfectly fine with the idea that there is not just one man, but
> 10,000 or many more men already in place on different mesas that are
> in fact selectably beneficial. In fact, there may be 10,000 or more
> men on each of 10,000 mesas. That is all perfectly fine and happens
> in real life. When something new "needs to be found", say, "vanilla
> with a cherry on top" or any other potentially beneficial function at
> that level of complexity or greater (this is not a teleological
> search you know since there are many ice-cream cones available), all
> of the men may search at the same time.

First, we seem to differ on the meaning of the "blind man". You seem to
be thinking of it as an organism. I am thinking of it as a modal
sequence in a population. That I am putting my modal sequence (there
will be continual mutations producing variants from this modal sequence,
but change in the modal sequence is determined by selection (which is
fast when it occurs) or neutral drift (which is slow but more frequent).
Neither zooms around the mesa, but they do explore it and also keep
the blind man on the top until or unless the geography changes.

But the math still proposes that these blind men are placed in random
spots away from the desired sequence and must search all of sequence
space to find the desired sequence.

And I beg to differ. You choose an end point and claim that some
average or random sequence must end up there and nowhere else. I see a
landscape filled with different mesas, each requiring a different
cluster of sequences for quite different functional results. A randomly
chosen blind man is not in a landscape that is flat except for the
teleologic goal you decided upon. The sequence space landscape of a
cell certainly has many (as many as 30,000-70,000 in humans -- think
gene number) mesas and presumably many more other unused or potential
mesas (sets of sequences) for functions that, at the present time, have
no selective value in humans. The landscape you describe is devoid of
contour except for a telephone pole at the teleologic goal. I would
expect a randomly chosen blind man that is not already on a mesa to walk
until it came across the first unoccupied slope available to it. Such a
slope would only exist when there is a selectable function that is not
currently occupied. It would climb that slope. Odds are that any such
slope leading to function will not be randomly distant. It will be
close by.


> My math certainly does not and never did presuppose that only one man
> may search the sequence space. That is simply ridiculous.

I find it ridicuolous that your math always assumes that the starting
point of any successful finder of the desired mesa must be, on average,
at an average or random position in sequence space. My position is that
out of all the useful (or even useless) sequences that do exist in a
cell at any given time, some will, even if only by chance, be much
closer to the desired mesa than others and *especially* much closer than
the average or random position. My point is that the probability is
that any *successful* random walk will most likely be from one of these
closest postion than from some average or random position. That means
that all your determinations of average distance has no relevance to any
successful walk. Only the positions of outliers close to the goal count.

> All the
> men search at the same time (millions and even hundreds of billions
> of them at times). The beneficial sequences are those sequences that
> are even slightly better than what is currently had by even one
> member of the vast population of blind men that is searching for
> something new and good.

Well, I would require significantly better in my simple algorithm.
Slightly better may or may not be significant.

> Now, if the average density of something new and good that is even
> slightly selectable as new and good is less than 1 in a trillion
> trillion, even 100 billion men searching at the same time will take a
> while to find something, anything, that is even a little bit new and
> good at the same level of specified complexity that they started
> with. On average, none of the men on their various mesas will be very
> close to any one of the new and good mesas within the same or higher
> levels of sequence space if the starting point is very far beyond the
> lowest levels of specified complexity.

Again with this "lowest levels of specified complexity" bullshit
verbiage. How do you measure this? The above is meaningless mantra to
avoid clear thought.

>> My presumption is that the successful search is almost always going
>> to start from the pre-existing mesa
>
>
> Agreed.
>
>
>> with the closest flavor to the new need (or from a duplicate,
>> which, as a duplicate, is often superfluous and quickly erodes to
>> ground level in terms of its utility).
>
>
> This is where we differ. Say you have chocolate and vanilla.
> Getting to the different varieties of chocolate and vanilla is not
> going to be much of a problem. But, say that neither chocolate nor
> vanilla are very close to strawberry or to each other. Each cluster
> is separated from the other clusters by thousands of miles. Now,
> even though you already have two clusters in your population, how are
> you going to evolve the strawberry cluster if an environmental need
> arises where it would be beneficial?

What makes you think evolution does this? To do that by the mechanism
of many single steps you would be starting out with a *random* sequence
relative to the end sequence. Now it *can* happen that large gaps can
be crossed (vide nylonase or the formation of chimeric proteins by
duplication). But these examples did not happen by the mechanism of a
trip of a thousand steps, changing one nucleotide at a time. These
examples happened in one swell foop by a single mutational event
involving changes in many nucleotides all at once to jump the sequence
close to or within where the slope goes up the mesa.

Some changes in what you might call 'complexity' do involve two
pre-existing subsystems that have independent utility combining to form
a new structure with a different utility. Usually the combining does
not involve many thousands of changes, but only those that cause
association between proteins to occur.

> You see, you make the assumption that just because you start out with
> a lot of clusters that any new potentially beneficial sequence or
> cluster of sequences will be fairly close to at least one of your
> 10,000 starting clusters.

Yes. Because the starting clusters are not sequences devoid of function,
but sequences that already perform biologically useful functions.

> This is an error when you start
> considering levels of sequence space that have very low overall
> densities of beneficial sequences.

How do you determine this, again?

> No matter where you start from
> and no matter how many starting positions you have to begin with,
> odds are that the vast majority of new islands of beneficial
> sequences will be very far away from everything that you have to
> start with beyond the lowest levels of functional complexity.

And how is it possible for every place in sequence space to be equally
far away from the new islands of beneficial sequences? Is the
beneficial islands in the center of a sphere and all other sequences on
the surface? Or perhaps the desired sequence is on one side of a Mobius
strip and all other sequences are on the other side. The geometry of
this is most intriguing. Could you describe a plane of sequence space
where all possible alternative sequences will be very far away from the
desired sequence?

And I have no way of determining how "levels of functional complexity"
fit into this plane of sequence space. Those words seem to have no
meaning whatsoever, except that they are used by you as a mantra to ward
away what you think is evil.


>
>
>> As mentioned, these pre-existing mesas are not random pop-ups.
>> They are at the most useful places in sequence space from which to
>> try to find near-by mesas with closely-related biologically useful
>> properties because they already have biologically useful
>> properties.
>
>
> Yes, similar useful biological properties would all be clustered
> together under one type of functional island of sequences. However,
> the overall density of beneficial sequences in sequence space
> dictates how far apart, on averages, these clusters of clusters will
> be from each other. New types of functions that are not so closely
> related will most certainly be very far away from anything that you
> have to start with beyond the lowest levels of functional complexity.
> You may do fine with chocolate and vanilla variations since those are
> what you started with, but you will have great difficulty finding
> anything else, such as strawberry, mocha, caviar, etc . . .

So when do you think this sort of leap from strawberry to mocha must
occur? Oh, I know. It must occur whenever you perceive the change must
involve something "beyond the lowest levels of functional complexity".
But those are just meaningless words.


>
> The suggestion that absolutely all of the clusters are themselves
> clustered together in a larger cluster or archipelago of clusters in
> a tiny part of sequence space is simply a ludicrous notion to me -
> outside of intelligent design that is. Oh no, you, Robin, Deaddog,
> Sweetness, Musgrave, and all the rest will have to do a much better
> job and explaining how all the clusters can get clustered together
> (when they obviously aren't) outside of intelligent design.

Not all sequence space is biologically useful. The clusters that exist
in cells are biologicially useful.


>
>
>> I *do* expect to see clustering in useful sequences. And I *do*
>> see it.
>
>
> So do I. Who is arguing against this? Useful sequences are often
> clustered around a certain type of function. What I am talking about
> is evolution between different types of functions.

The change in function from enzyme to lens crystallin, for example?
Wasn't that a change between two quite different types of function?

> The evolution of
> different sequences with the same basic type of function is not an
> issue at all. It happens all the time, usually in the form of an
> up-regulation or down-regulation of a certain type of function, even
> at the highest levels of functional complexity.

Or changing substrates? Or connecting a pre-existing neural pathway to
a change in rhodopsin? Or keeping substrates but changing enzymatic
activity? Or binding two proteins together in a specific stoichiometry?
Which of these is a change in "the highest levels of functional
complexity"? And could you show the math that allowed you to identify
the change as one at "the highest levels of functional complexity"?

> But, this sort of
> intra-island evolution is a far cry from evolving a new type of
> function (i.e., going from one cluster to another). In fact, this
> sort of evolution never happens beyond the lowest levels of
> functional complexity due to the lack of density of beneficial
> sequences at these higher levels of specified complexity.

Why is the density of beneficial sequences any lower or higher dependent
upon the "level of specified complexity"?

david ford

unread,
Jan 16, 2004, 11:52:16 PM1/16/04
to
Sean Pitman <seanpi...@naturalselection.0catch.com> in
"Re: Is Evolution an Anti-God Theory?" on 1 Jun 2003:

> It seems to me that the concept of IC is quite helpful indeed. The
> problem is that many, even Behe himself, seem to try to limit the
> definition of IC to "very complex" systems of function in order to
> show that IC systems cannot evolve.
>
> As I see it all systems of function are IC. It is just that some
> systems are more simple than other systems of function. There is a
> spectrum of complexity, but all systems along this spectrum from
> simple to more and more complex are all IC. In other words, not all
> setups of a given number of parts or part types will be able to
> perform a given function. The parts in any system of function can in
> fact be altered, removed, or ordered in a different manner so that the
> function of the system is completely destroyed. In fact, there are
> vastly more non-functional potential arrangements of parts than there
> are beneficially functional arrangements of parts in a particular
> scenario.

Compare Dawkins on ways of being dead
http://tinyurl.com/2aov5
aka
http://www.google.com/groups?selm=Pine.SGI.3.96A.990406232938.942967A-100000%40umbc8.umbc.edu

> Take, for example, Behe's famous mousetrap IC illustration. Many try
> to argue that a mousetrap is not IC since parts can be removed or
> changed and it still can catch mice. That is not the issue. If you
> change the mousetrap, it may still catch mice, but not in the same
> way. The changed mousetrap is a different mousetrap that catches mice
> in a different way. Certainly there are many different kinds of
> mousetraps that can catch mice, some more effectively than others.
> However, all of these mousetraps are dependent upon a certain number
> of parts that are all arranged in a very specific way in order for
> these parts to work together to catch mice (i.e., To perform their
> function). Clearly there are a lot more arrangements of mousetrap
> parts for any given type of mousetrap that would not catch mice at
> all. All mousetraps can in fact be reduced or changed in a way that
> would destroy their function completely. And, these potential
> non-functional mousetraps are far more numerous than those
> comparatively few arrangements than can actually perform the mouse
> catching function.
>
> Of course, it is theoretically possible to arrange several of these
> working mousetraps in sequential order so that very small steps seem
> to exist as one moves from one type of trap to the other. Obviously
> then, it is NOT impossible for IC systems to evolve via function-based
> selection mechanisms since such an evolutionary path need not
> necessarily cross wide neutral gaps in function or non-function. The
> problem is that these gaps are often wider than one might initially
> think.
>
> http://naturalselection.0catch.com/Files/irreduciblemousetrap.html
>
> Even functions that are based on the workings of single proteins, such
> as the enzymatic functions of lactase or nylonase, are IC in that
> there are a limited number of parts that are required to give rise to
> that particular type of function. For more simple functions, such as
> these single-protein-based functions, there might be a much higher
> ratio of sequences of a given length or smaller that would be able to
> perform a given function, like the lactase or nylonase function. For
> example, given a sequence of amino acids 1,000aa in size, there are
> about 1 x 10e1300 different possible protein sequences. This is an
> absolutely huge number of different possibilities. It is a 1 with
> 1,300 zeros following it. Out of all of these possibilities, how many
> would have the lactase function? Certainly there would be many of
> these sequences that would have the lactase function, but certainly
> not all of them or even most of them. Perhaps the ratio would be as
> high as 1 in a trillion? If the ratio where 1 in a trillion, that
> means that any given functional lactase AA sequence would be
> surrounded by an average of 1 trillion non-lactase sequences. If a
> particular functioning lactase sequence is changed or "reduced" beyond
> a certain point, it will no longer function at all, not even a little
> bit. This is the definition of IC. The lactase function, even though
> based in the AA sequence of a single protein, is IC. Of course,
> compared to other systems of function, the lactase and nylonase single
> protein enzymes are not all that complex since there are is a
> relatively high percentage of potential lactase sequences as compared
> with the total number of possible sequences out there. Because of
> this, these functions are relatively simple, requiring a relatively
> short stretch of DNA to code for their function. Other systems of
> function require multiple proteins all working together
> simultaneously. Much more DNA real estate is necessary.
>
> Before thinking about more complex systems function, such as bacterial
> motility, consider that even the evolution of the relatively simple
> lactase function is quite difficult. Barry Hall demonstrated this in
> several experiments where he deleted the lacZ genes in E. coli
> bacteria to see if they would evolve the lactase function back again
> using some other genetic sequence. And, they did evolve the lactase
> function in just one or two generations. As it turned out, a single
> point mutation to a completely different DNA sequence was able to
> produce a selectively advantageous lactase function in a lactose
> environment. Hall called this "evolved" sequence the ebg gene
> (evolved beta galactosidase gene). But, he started wondering, "If
> this worked for the deletion of the lacZ gene, what will happen if I
> delete the ebg gene too?" So, Hall deleted the ebg and lacZ genes in
> certain colonies of E. coli. What happened next is very interesting.
> These double mutant E. coli colonies never evolved the lactase
> function back again despite high population numbers, high mutation
> rates, 4 million base pairs of DNA each, positive selection pressure,
> and tens of thousands of generations.
>
> http://naturalselection.0catch.com/Files/galactosidaseevolution.html
>
> Now, why didn't Hall's double mutant E. coli colonies evolve the
> relatively simple lactase function back again? Hall himself described
> this colonies as having, "limited evolutionary potential." What was
> it that limited their ability to evolve the relatively simple lactase
> function despite very positive benefits if they were to ever evolve
> this helpful function?
>
> It seems that neutral gaps existed between what was there and what was
> needed. The genetic real estate of this huge population of E. coli
> simply was not large enough to undergo the random walk across this
> neutral gap in beneficial function despite being given thousands of
> generations.
>
> Obviously then, even such simple functions as the function of single
> proteins are IC and this can and often does create difficulties for
> mindless evolutionary processes. The problems only increase
> (exponentially) as one moves up the spectrum of complex systems.
>
> Sean
> www.naturalselection.0catch.com

Von Smith

unread,
Jan 17, 2004, 2:33:09 AM1/17/04
to
drea...@hotmail.com (Von Smith) wrote in message news:<8d74ec45.04011...@posting.google.com>...

> seanpi...@naturalselection.0catch.com (Sean Pitman) wrote in message news:<80d0c26f.04011...@posting.google.com>...
> > "Chris Merli" <clm...@insightbb.com> wrote in message news:<lKmNb.55023$5V2.67607@attbi_s53>...

<snip>

>
> I would have been more impressed if you had written this *after*
> giving a substantive reply to Deaddog's recent excellent post on
> Synthetic Biology, which probably sheds some light on how biologists
> *really* think complex multi-protein systems might evolve. In it, he
> cites a paper in which researchers randomly switched around some of
> the parts involved in complex multi-protein interactions to see what
> they would do.
>
> Combinatorial synthesis of genetic networks.
> Guet CC, Elowitz MB, Hsing W, Leibler S.
> Science. 2002 May 24; 296(5572): 1466-70.
>
> http://www.sciencemag.org/cgi/content/full/296/5572/1466
>
> So what happens when one shakes up the regulatory bits of a biological
> system and lets them fall where they will? AIUI, far from ending up
> with nothing but random junkpiles, the researchers were able to obtain
> a variety of novel logically-functioning phenotypes. No need for some
> pre-existing homonculus magically prompting the various parts on how
> to behave: as often as not the parts were able to associate and
> interact coherently left to their own devices. Of course it is
> possible that this liberal arts major is misunderstanding the article.
> Perhaps the biologically washed can comment more coherently.
>

To be fair and accurate, I should note that the researchers did not
actually shuffle the constituent parts of their combinatorial
libraries randomly; they took a sequence of three transcriptional
regulators, and associated each one with any one of five promoters.
This yielded a "sequence space" of 125 possible arrangements of
promoters and regulators. With a combinatorial library this small,
Guet et al didn't have to sample the population, they were able to
survey all the possible combinations in the library. The point is
that, out of all these possible combinations, there was not an
overwhelming majority of junk piles with maybe two or three working
combinations, which Dr. Pitman might expect to be the case, but rather
a goodly proportion of the combinations worked coherently. In other
words, if they *had* shuffled the promoters randomly, the odds that
they would have ended up with a coherent complex system would actually
have been quite good.

Von Smith

unread,
Jan 17, 2004, 4:12:43 AM1/17/04
to
seanpi...@naturalselection.0catch.com (Sean Pitman) wrote in message news:<80d0c26f.04011...@posting.google.com>...

> howard hershey <hers...@indiana.edu> wrote in message news:<bu46sv$srt$1...@hood.uits.indiana.edu>...
>

<snip lengthy discourse in which the visually impaired are compelled
to wander dessert wastelands in search of ice cream mesas>

>
> And what about the other clusters? Is the environment going to change
> just right a zillion times in a row so that bridges can be built to
> the other clusters?

<snip some more>

I wanted to highlight a couple of issues that Dr. Pitman raises at the
end of this rather lengthy post. His argument here, if I understand
it correctly, is that: granted that useful sequences of a *given*
function may tend to form clusters in a sequence space, rather than
being scattered randomly and sparsely throughout it, it is nonetheless
the case that various clusters of sequences serving *different*
functions will tend to be thus widely spaced apart, so that a sequence
sitting atop a mesa of mint chocolate chip ice cream cones is likely
to be far away from any other mesas of other given flavor, such as
butter pecan. Our visually challenged hero may be able to find other
varieties of bubble gum fairly easily, but the chances of our his
successfully traversing the Rocky Road to a different flavor is
vanishingly small.

Now, I suppose one way to determine if this is actually the case is to
wave one's hands (being careful not to drop that triple scoop of mint
chocolate chip) and make guesses based on personal incredulity and
weak analogies; this seems to be Dr. Pitman's preferred method of
inference and argument. Another possibility might be to see if what
we actually know about the real world could provide us with some
clues.

First, it is important to note that the underlying issue here is
whether the various complex structures we observe in life today might
have evolved from different complex structures that existed in life
yesterday. This is *not* the same question as whether any two
functions we might select at random are likely to be able to evolve
into one another. It may very well be nearly impossible for a
flagellum to evolve from, say, a Golgi apparatus or a 2,4-DNT enzyme
cascade. even given "zillions" of years. Who cares? No one is
suggesting such a thing. Structures such as Tsp pili or TTSS weren't
proposed as possible flagellum precursors at random. They interest
biologists because of actual evidence that there are significant
homologies between them and the flagellum.

To reiterate: we are not randomly selecting any two functions and
discussing whether or not they might have evolved; our sampling method
is quite biased. Now Dr. Pitman has objected on several occasions
that he is not assuming that such structures evolve from random
sequences, either. He may well be sincere in this. It may well be
that he just does not appreciate how his arguments and claimed
probability calculations are sensitive to this assumption.

So just how widely scattered are the different mesas of functions in
life today from one another, and from the different mesas of function
that existed in life yesterday and 500 Mya?

Well, my casual survey of different functions in life tells me that,
quite often, dramatically different functions can be served not only
by closely-related clusters of mesas; they can often be found on the
*mesa*. Consider the following functions:

improving flight;
display;
insulation;
camouflage.

The exact same bird feathers perform *all* of these functions. Not
only that, but the same protein that goes into making these feathers
also goes into making the bird's beak, and is in turn essentially the
same protein found in mammal hair and reptile scales. Several
functions stacked up on the exact same mesa.

But, one might argue, structures like feathers are relatively simple.
As one goes up the ladder of complexity, however, etc. etc. Well, OK,
let's take the whole suite of adaptations that make most birds
efficient fliers, including wing shape, pneumatic bones, a
super-efficient respiratory system, etc. How likely is this complex
system of adaptations for flight to be adaptable for a different
function? There's no need to wonder, as we have actual evidence.
Penguins, with more or less the same suite of adaptations, use them to
swim instead of fly (and of course, some birds are quite proficient
both as swimmers *and* fliers).

Note that it wasn't necessarily obvious ahead of time that swimming
and flying adaptations would overlap. AFAIK, bats aren't very good
swimmers at all, nor are butterflies. Likewise, I don't expect
whales, squids, sharks, or sea tortoises, for all their swimming
prowess, to be especially good fliers, although I suppose there are
"flying" fish. I don't expect that pterodons evolved from pleiosaurs,
or vice versa. In a world without penguins or cormorants, I'm not
sure how useful our prior intuitions would be about how likely
different functions are to overlap. One has to compare actual
*structures*, and try to determine whether a reasonable evolutionary
pathway exists between a structure and some *logical* precursor, not
some randomly selected one.

OK, that was on the level of gross animal analogy. What about at the
molecular level? We already know about enzymes doubling as lens
crystallins, immunoglobins with enzyme activities, Hox genes that code
for body segmentation in one organism and hindbrain formation in
another. Blah blah blah. But of course those are relatively simple,
closely-related functions, so they don't really count. We all know
that as we move up the ladder of complexity, these sorts of
overlapping clusters of differing functions go away, or at least
become vanishingly rare, right?

Let's see. Biologists propose that a highly complex motility function
might have evolved from an ancestral secretory system. What are the
odds, one might ask, that a cluster of motility "mesas" was likely to
be somewhere close enough to a cluster of secretory "mesas" to evolve
from it? Well, using the Incredulous Hand-waving, Weak Analogy
Method, we can rigorously calculate that they are so far away from one
another that our blind ice cream enthusiast couldn't possibly travel
from one mesa to the next in less than 17.8 zillion years.

Surprisingly, however, the Actually Look At the Evidence Method yields
a dramatically different result: it turns out that the flagellum
actually *is* also a secretory system. Not only are the clusters of
mesas not widely scattered, they overlap. At least one of our blind
man's ice cream cones already has more than flavor stacked on top of
it.

Further application of the ALAE method yields even more discrepancies
with the IHWA method: when we do phylogenetic analyses of genes to
determine where they *do* actually cluster, we find gene families and
super-families that encompass a variety of different functions, while
the set of genes serving the same function might be scattered among
different familes or "clusters" of genes.

So what assurance do we have, Dr. Pitman might demand, that no matter
where our blind heroes start out, that at least some of them will
always find a way to get their cherry cheesecake fix? None
whatsoever. And in fact the current lowlands of our dessert landscape
will be littered with the bleached bones of failed Cookies & Cream
afficionados. It may well be that the overwhelming majority of all
clusters of *possible* functions is out in the uncharted territory
that our natural history will never see. But we don't care about
those. Are problem is simply to locate the clusters of functions that
actually exist, and to compare them one another, and to the clusters
of functions that actually existed yesterday. When we do this, we
find evidence of precisely the sort of inter-function clustering that
Dr. Pitman finds unlikely.

The Sandman has given me a past due notice, so I will stop for now.

Von Smith
Fortuna nimis dat multis, satis nulli.

It m

Zachriel

unread,
Jan 17, 2004, 2:41:06 PM1/17/04
to
"Sean Pitman" <seanpi...@naturalselection.0catch.com> wrote in message
news:80d0c26f.03120...@posting.google.com...
> So, what you
> "start with" is quite important to determining what is and what is
> not beneficial. Then, beyond this, say you start with a short
> sequence, like a two or three-letter word that is defined or
> recognized as beneficial by a much larger system of function, such as
> a living cell or an English language system. Try evolving this short
> word, one letter at a time, into a longer and longer word or phrase.
> See how far you can go. Very quickly you will find yourself running
> into walls of non-beneficial function.

First you made this challenge. I responded with a word puzzle where,
starting with the single letter word "O", and by only changing one letter at
a time, and with concatenation, I constructed the phrase, "Beware a war of
words, Sean Pitman, ere you err."


"Sean Pitman" <seanpi...@naturalselection.0catch.com> wrote in message

news:80d0c26f.04011...@posting.google.com...
<snipped>

> For example, start with a meaningful English word and then add to or
> change that word so that it makes both meaningful and beneficial sense
> in a given situation/environment. At first such a game is fairly easy
> to do. But, very quickly you get to a point where any more additions
> or changes become very difficult without there being significant
> changes happening that are "just right". The required changes needed
> to maintain beneficial meaning with longer and longer phrases,
> sentences, paragraphs, etc., start to really get huge. Each word has
> a meaning by itself that may be used in a beneficial manner by many
> different types of sentences with completely different meanings.
> Although the individual word does have a meaning by itself, its
> combination with other words produces an emergent meaning/function
> that goes beyond the sum of the individual words. The same thing
> happens with genes and proteins. A portion of a protein may in fact
> work well in a completely different type of protein, but in the
> protein that it currently belongs to, it is part of a completely
> different collective emergent function. Its relative order as it
> relates to the other parts of this larger whole is what is important.
> How is this relative order established if there are many many more
> ways in which the relative order of these same parts would not be
> beneficial in the least?
>

Now you have upped the challenge to longer sentences and paragraphs. This
may surprise you, but it is much much easier to make longer words, sentences
and paragraphs once we have a starting library of phrases. As you point out,
the same word has many different uses. This is very similar to biology
whereby a feather can insulate as well as provide lift for flight, or where
limbs can be adapted for walking, flight or swimming.

Despite your incredulity, Sean, it is quite easy to construct sentences,
paragraphs, even whole essays from these simple rules. You must try to keep
in mind that ignorance is not evidence.

To make the exercise a little more challenging, I have adopted a loose
iambic pentameter. All words are found in Merriam-Webster, excepting Sean's
own name. Some of the nonsense verse is quite interesting, "like, lick,
lock, block, click, clock, slick, stick, stack" or when concatenating
phrases, "a war, of words, a war of words, beware a war of words"

But now we are after whole lines and verses. Here we go . . . (complete poem
at the bottom of post) . . .

-----------------------

Rules: Change only one letter at a time from any existing string. Can
concatenate any two strings. However, only one operation at a time. All
words, phrases and sentences must make sense in standard English.

Starting with a single letter word, "O".

----------------------------------

o, a, i
o, or, ore, one, wore, word, whore

words, wordy, ward, war, tar, wars, ware, tare, are
ere, err, era, ore, ode, of, off, or, our, your, you
ire, irk, irks, lire, lyre, fire, lice, like, lick,
lock, block, click, clock, slick, stick, stack
for, fore, form, forms, foreword
ow, row, brow, prow, prom, from

war, wan, man, may, mean, many, bean, bear, beer, bee
be, ear, year, dear, tear, pear, spear, dean, deal, ideal, idea
sean, sear, bead, lead, seer, steer, steed, stead
eat, ear, seat, set, wet, we, see, sit, pit, it, is, in, gin, instead
seep, step, pet, poet, poem

ion, sir, stir, stair, staid, tee, tea, teat, tear
treat, great
as, an, can, and, ass, pass, piss, kiss
to, do, so, go, no, not, nod
sin, tin, kin, king, win, wine, pine, pin, ping
wee, weeping, weening, weaning, meaning
is, his, this, him, hem, he, the, thy, why, who, thin, think, thing

----------------------------------

be-ware
wordy ward, a wordy ward, word wars
a war, a kiss, world, world war

of words, a war of words
beware a war of words
pit-man, Sean pitman
beware a war of words Sean pitman
you err
I err
ere you err

a war, of words, a war of words
beware a war of words
pit-man
Sean pitman
beware a war of words Sean pitman
you err
I err
ere you err

* Beware a war of words, Sean Pitman,
* Ere you err.

----------------------------------

piss, puss, pus, bus
but, jut, just
O, ow, row, crow, crown, crowd
crew, grew, grow
lo, low, lowe, lower, lowers, slow, log, blog
do, doe, dose, lose, rose
loss, close, chose, choose
not, snot, soot, shoot, hoot
me, some, same, tame, time
sometime, sometimes

do, doe, does, dole, pole, bole, bold, old
cold, could, would
no, now, know, known
err, error
ass, ash, lash, clash

O Sean Pitman,

elk, elm, helm, hell, well, help
elf, self, shelf
at, cat, hat, that, hate, have
ate, rate, crate, create, grate

rat, ray, tray, stray, astray
swords

A man, A man wins, the crown, A man wins the crown
is helm, lowers his helm, but lowers his helm
A kiss, is a kiss, a war, be just, can be just, a war can be just
the crowd, irks the crowd, just irks the crowd
leads you, leads you well, leads you well astray
you know, can lead, a clash, of swords
a clash of swords, to a clash of swords.

* A man wins the crown, but lowers his helm. A kiss
* Is a kiss, and a war can be just, but a war of words
* Just irks the crowd and leads you well astray.
* Words, you know, can lead to a clash of swords.

----------------------------------

fa, got, fagot, faggot
rag, lag, leg, it, tit, legit, sag, sage, sages

eve, ever, every, aver, eye, eyes
every-one
not, her, nother, another, other, others
me-me, meme, mere

hat, what, whet, whey, why, when
ate, late, hate
one, lone, alone
kin, kiln, kill, till, tall, all, kind, find
sword, sworn, shorn
i, id, lid, did

O Sean Pitman,

you think, do you think, you alone, have it
why do you think, that you alone, have it legit
when others, aver another, aver another idea

* Why do you think that you alone have it
* Legit when others aver another idea?

----------------------------------

rat, rot, lot, slot, slut
ink, pink, oink, oink
i, bi, by, got, bigot, bight, blight, light
big, bit, bite, byte
rig, orig, origin, origin
led, pled, bled, bleed
dim, dimpled, dimple, simple

Could it, could it be, you could, that you could
the light, see the light, that you could see the light
But choose, but choose instead
your eyes, close your eyes, and block
close your eyes, to close your eyes
the sight, block the sight
The origin, of life, we know
The origin of life, The origin of life we know
this poem, like this poem, just like this poem
simple forms, rose from simple forms
in meaning, in kind, and in kind, by step
step-by-step

* Could it be that you could see the light
* But choose instead to close your eyes and block
* The sight? The origin of life we know
* Just like this poem rose from simple forms
* In meaning, and in kind, step-by-step.

----------------------------------

We can trace the "etymology" of each word used in the poem. Some of the more
difficult words to create include "light", "choose", "instead" and "simple".

o, go, got, i, bi, bigot, bight, light
o, do, doe, dose, lose, close, chose, choose
i, in, o, or, ore, ere, err, ear, sear, seer, steer, steed, stead, instead
i, is, his, him, dim, id, lid, led, pled, dimpled, dimple, simple

(It would have been much easier if we had allowed prefixes and suffixes,
like free radicals in chemistry; and instead of merely letters, had included
phoenetics, such as "sh" or "tr"; or allowed letter rotations; or allowed
dropping letters when concatenating; but that would have been much too
easy.)

----------------------------------

* Beware a war of words, Sean Pitman,
* Ere you err. O Sean Pitman hear me:

* A man wins the crown, but lowers his helm. A kiss
* Is a kiss, and a war can be just, but a war of words
* Just irks the crowd and leads you far astray.
* Words, you know, can lead to a clash of swords.

* Why do you think that you alone have it
* Legit when sages aver another idea?

* Could it be that you could see the light
* But choose instead to close your eyes and block
* The sight? The origin of the life we know
* Just like this poem rose from simple forms,
* In meaning, and in kind, step-by-step.

Uncle Davey

unread,
Jan 17, 2004, 3:47:52 PM1/17/04
to

"Zachriel" <an...@zachriel.com> wrote in message
news:100j472...@corp.supernews.com...

Very clever.

Uncle Davey


Chris Krolczyk

unread,
Jan 17, 2004, 4:40:42 PM1/17/04
to
dfo...@gl.umbc.edu (david ford) wrote in message news:<b1c67abe.0401...@posting.google.com>...

(huge snip)

That's nice, David. Other than the typical self-refential URL
and a completely redundant quoting of Pitman, what's your point?

-Chris Krolczyk

howard hershey

unread,
Jan 19, 2004, 11:49:00 AM1/19/04
to

Sean Pitman wrote:

Let's call this the Charlie Wagner ploy. When a creationist tires of
claiming that the assembly of protein structures is too complicated and
God must have, therefore, done it, he/she/it/they then turns to DNA as
the mysterious intelligent entity that encodes and *enacts* the wisdom
of the ages. They, the masters of bad analogical arguments, analogize
the DNA of a cell as its 'brain', intelligently directing everything the
cell does, including determining what gets transcribed and how proteins
assemble into structures like flagella.

Alas, DNA's *only* contribution to the assembly of protein structures is
*encoding* the protein's amino acid sequence and *short* regulatory
sequences. DNA does not *enact* anything by itself. Genetic
information does not, by itself, tell where, when, or how much of each
part of a system is made. The cell's proteins, interacting with and
sensing the environment, interact with each other (in long chains and
cascades of reactions) to modify DNA-binding proteins so that they
either bind or release particular *short* DNA sequences. The *short*
regulatory (DNA-binding) sequences are well under the "hundreds to
several thousands" of changes you say evolution cannot produce. They are
typically 6-10 nucleotides in length, since that is the number of
nucleotides that can be seen in a single helical twist in the major
groove of DNA. These sequence elements can then allow a response to
environmental cues by either allowing or discouraging the formation of
the protein complex called RNA polymerase to make or not make a mRNA
transcript of this sequence infomation. [Notice that DNA is a passive
recipient of actions performed upon it.] That's it. I am not saying
that the sequence information in DNA is unimportant. I am saying that
that is *all* that the DNA provides. DNA is a dumb, unintelligent
molecule. DNA is *acted upon* by the cell in response to environmental
cues by the action of the cell's proteins. DNA does not act
independently as an intelligent agent in any way. BTW, *because* the
regulatory sequences of DNA are so *short* and have so little complex
information, it is not surprising that much of evolutionary change
involves changes in regulation rather than change in sequence. It is
quite possible and very easy to change regulation by random mutation
producing a new regulatory region or by combinatorial changes putting a
gene under new regulatory regions. Witness the eye crystallins.

Let's go through some very basic knowledge about DNA, RNA, and protein
that Sean has seemingly never learned, or if learned, only for
responding on a test rather than for understanding. These basic ideas
are so fundametntal they are even called The Central Dogma.

1) DNA gets transcribed by RNA polymerase to produce an mRNA.
Regulatory proteins that are responsive to environmental cues interact
with regulatory sequences to initiate this process (often via a chain or
regulatory cascade of enzymes). The DNA dumbly and stupidly responds to
these environmental cues by transcribing a mRNA when the proper proteins
are in the proper position. That DNA responds dumbly rather than with
intelligence is shown by the fact that moving a regulatory sequence from
its original postion in DNA or by creating a regulatory sequence
elsewhere by mutation results in transcription from that new position
despite the fact that doing so might be utterly without value to the
cell and not what the cell needs. Molecular biologists take advantage
of this 'stupidity' of DNA by creating hybrid chimeric molecules
(chimeric molecules also happen in nature) with, say, a bacterial
florescent protein, under the control of the regulatory sequence of the
insulin gene. They do this so that they can literally *see* when the
insulin genes are being transcribed. Transcription is not an
intelligent process. It is a dumb chemical process. DNA is acted upon
by proteins. It only encodes information. It is not an independent
actor. By itself, DNA does nothing. It is only useful as a part of a
system.

In eucaryotic cells, after the mRNA is transcribed it is typically
further processed to remove introns, add 5' caps, and 3' tails. [In
bacteria, these processes are missing.]

Then the processed mRNAs then go out out of the nucleus where they are
translated into proteins. In bacteria, translation starts immediately
after transcription and is proceeding even as transcription continues.

But DNA, neither the genes being encoded nor DNA as a general molecule,
plays any role in any of these steps (other than providing the mRNAs
and, via the same process, the proteins that provide these functions).
In eucaryotes, in particular, there is always a nuclear membrane between
DNA and the translation machinery.

The direct role of DNA in whatever a protein does extends *only* to the
point of producing the primary transcripts. Any subsequent effect is a
consequence of the sequence of that primary transcript.

Translated proteins then are often clipped, chaparoned, differentially
transported or otherwise modified during or after translation. The DNA
that encodes this protein does nothing in any of these steps.

The proteins then aggregate with one another due to the fact that they
have sites that cause them to attach more or less strongly to each
other. Environmental conditions (such as concentration of protein, the
presence of a seed protein, the level of O2, the level of Ca, the
presence or absence of small allosteric effectors -- small molecules
like sugars or cAMP that change the conformation of protein when bound)
may differentially influence whether two or more proteins assemble with
sufficient binding strength or dissassemble. Sequence of the protein,
again, is the only thing that DNA does that affects assembly of proteins
into higher order structures. Notice that the encoding DNA is not even
present anywhere *near* where the proteins self-assemble into their
final structure. So how does Sean imagine the DNA directing the
self-assembly of flagella? By ESP? By neuronic tentacles?

Examples: Sickle cell anemia and normal hemoglobin affect the structure
and shape of the entire rbc. The change in conformation in low O2 of the
sickle cell is due to changes in environmental conditions in the
complete absence of any nuclear DNA in the cell (since mammalian cells
are enucleate). Ribosomes can self-assemble in a test tube. So can
entire viruses, often one can encapsulate non-viral DNA or only a short
sequence (acted upon by proteins) of viral DNA. That is, the role of
DNA in phage assembly has little or no relationship to its sequence.

Mitotic spindles can be made to assemble into long tubes or dissassemble
by merely changing environmental conditions (Ca ion concentration plays
a big role). A change in structure in a particular protein, because of
the presence of a 'seed' protein, can make a cow very, very angry. This
can happen without any change in the DNA or in protein synthesis.
Environmental conditions (including environmental conditions that change
transcription rates by feedback through proteins) also regulate the
construction of the sub-parts of the bacterial flagella. DNA is not
involved in the assembly of a single part of the bacterial flagella
*except* indirectly as it affects the sequence of the proteins. DNA
supplies the raw materials, the proteins whose sequences allow them to
self-assemble, but their assembly into higher order structures is
entirely independent of DNA.

> Without this pre-established information the right parts just
> won't assembly properly beyond the lowest levels of functional
> complexity. It would be like having all the parts to a watch in a
> bag, shaking the bag for a billion years, and expecting a fully formed
> watch, or anything else of equal or greater emergent functional
> complexity, to fall out at the end of that time. The same is true for
> say, a bacterial flagellum. Take all of the necessary subparts needed
> to make a flagellum, put them together randomly, and see if they will
> self-assemble a flagellar apparatus.

In fact, the *proteins* of the bacterial flagella do self-assemble in a
particular order. It is that ontologic order that tells us how the
flagella likely arose via specific subsystems that were *independently*
derived and subsequently co-opted to perform their present function.
For example, the L and P ring proteins reach their positions
independently (via a different export machinery) from the TTSS export
machinery that pumps out all the closely-related flagellar proteins. The
TTSS export machinery of the flagella, of course, self-assembles first.
The motor assembles independently of the TTSS export machinery and then
becomes attached as a sub-system. So the flagella looks like the
assembly of independently useful (or potentially independently useful
and independently regulated) subsystems that were co-opted to perform a
new function rather than the assembly of a single system.

The base of the flagella is clearly capable of performing the
independent function of protein transport. It still does perform that
function. The motor and regulatory proteins clearly have independent
utility, as closely related proteins still perform these functions
elsewhere. The L and P rings also have relatives that serve similar
functions. The flagellar whip proteins are all structurally similar and
probably represent duplication and specialization events. But the
flagellar proteins all have the ability to bind to one another to form
the tube (when they reach the tip of the growing tube, but not before)
because of this sequence relatedness. Some of the specialization
probably is due to the fact that the first proteins are released from
the growing tip into a different environment than is the case for later
whip proteins. But there is no doubt at all that the flagellar whip
*self-assembles* and does so in the complete absence of the DNA that
informed its sequence. There is no doubt at all that, if we know the
right environmental cues, flagellar whip proteins could be induced to
form flagellar whips in a test tube in the absence of DNA.

> It just doesn't happen outside
> of the very specific production constraints provided by the
> pre-established genetic information that code for both flagellar part
> production as well as where, when, and how much part to produce so
> that assembly of these parts will occur in a proper way. The simple
> production of flagellar parts in a random non-specific way will only
> produce a junk pile - not a highly complex flagellar system.

Yes. Cells are systems. But no one but you is envisioning the flagella
evolving from some sort of junk pile of randomly produced proteins.
Rather, the flagella evolved by co-opting already useful systems to
perform an additional functionally useful activity.


>
> Now, of course, if you throw natural selection into the picture, this
> is supposed to get evolution out of this mess. It sort through the
> potential junk pile options and picks only those assemblages that are
> beneficial, in a stepwise manner, until higher and higher systems of
> functional complexity are realized. This is how it is supposed to
> work. The problem with this notion is that as one climbs up the
> ladder of functional complexity,

If there is a "ladder of functional complexity" I would like to know
about it. How is it determined that system A is higher or lower on this
ladder than system B? How do you imagine evolution working its way up
the ladder? By first magically poofing utterly useless pieces of junk
protein and then magically poofing all those pieces directly to the top
of the ladder? That seems to be your strawman du jour.

> it becomes more and more difficult to
> keep adding genetic sequences together in a beneficial way without
> having to cross vast gaps of neutral or even detrimental changes.
>
> For example, start with a meaningful English word and then add to or
> change that word so that it makes both meaningful and beneficial sense
> in a given situation/environment. At first such a game is fairly easy
> to do. But, very quickly you get to a point where any more additions
> or changes become very difficult without there being significant
> changes happening that are "just right". The required changes needed
> to maintain beneficial meaning with longer and longer phrases,
> sentences, paragraphs, etc., start to really get huge. Each word has
> a meaning by itself that may be used in a beneficial manner by many
> different types of sentences with completely different meanings.
> Although the individual word does have a meaning by itself, its
> combination with other words produces an emergent meaning/function
> that goes beyond the sum of the individual words. The same thing
> happens with genes and proteins. A portion of a protein may in fact
> work well in a completely different type of protein, but in the
> protein that it currently belongs to, it is part of a completely
> different collective emergent function. Its relative order as it
> relates to the other parts of this larger whole is what is important.
> How is this relative order established if there are many many more
> ways in which the relative order of these same parts would not be
> beneficial in the least?

Ever hear of transpositon, deletion, duplication? Those are mechanisms
that can bring together different functional parts. But keep in mind
that most evolutionary change is more a matter of change in quantity
(regulation) than in structure. After all there are very significant
few structural differences at the DNA level between humans and chimps.


>
> Again, just because the right parts happen to be in the same place at
> the same time does not mean much outside of a pre-established
> information code that tells them how to specifically arrange
> themselves.

Notice that in the model of evolution of the bacterial flagella that
zosdad has pointed you to (it was written by someone he is quite in
touch with) that all of the precursors had independent utility and
already self-assembled into independently useful sub-components.

>>>In
>>>order to keep up with this exponential decrease in average cone
>>>density, the number of blind men has to increase exponentially in
>>>order to find the rarer cones at the same rate. Very soon the
>>>environment cannot support any more blind men and so they must
>>>individually search out exponentially more and more sequence space, on
>>>average, before success can be realized (i.e., a cone or cluster of
>>>cones is found). For example, it can be visualized as stacked levels
>>>of rooms. Each room has its own average density of ice cream cones.
>>>The rooms on the lowest level have the highest density of ice cream
>>>cones - say one cone every meter or so, on average. Moving up to the
>>>next higher room the density decreases so that there is a cone every 2
>>>meters or so. Then, in the next higher room, the density decreases to
>>>a cone every 4 meters or so, on average. And, it goes from there.
>>>After 30 or so steps up to higher levels, the cone density is 1 every
>>>billion meters or so, on average.
>>
>>If the development of each protein started from scratch you may have an
>>excellent arguement but nearly all proteins from other proteins so you are
>>starting from a point that is known to be functional.
>
>
> You are actually suggesting here that the system in question had its
> origin in many different places. You seem to be suggesting that all
> the various parts found as subparts of many different systems somehow
> brought themselves together to make a new type of system . . . just
> like that.

No. The independent subsystems came together by several independent
steps, each useful in its own right. No one is suggesting that the
bacterial flagella appeared by the equivalent of a three-body collision.
We are suggesting that it arose by two two-body collisions with a
useful, albeit transient, intermediate. If you think it requires a
four-body collision to go from four independent subsystems to a final
single system, you are wrong there as well. It requires three two-body
collisions. You keep failing to notice that all of the subsystems
proposed (the TTSS function of the base, the regulatory functions of the
regulatory proteins, the motor functions of the mot proteins, the P and
L rings, and even the whip/injectosome) had independent useful functions
assigned to them and basically did not change their biochemical actions
to a major degree when they became co-opted into a flagellar system.
The flagella was not assembled out of "junk". It was assembled out of
subsystems (or duplicates thereof) that were co-opted into performing a
related biochemical action in service of a new developing function.
Subsequent specialization of these subsystems for this new function was
a *consequence* of the utility of the new function: because, in these
cells, motility was a useful selectable function, selection for changes
that improved that function was favored.

> Well now, how did these various different functional
> parts, as subparts of many different systems, know how to come
> together so nicely to make a completely new system of function?

They didn't. It was a process of trial and error. When a trial
produced a useful intermediate, it was kept.

> This
> would be like various parts from a car simply deciding, by themselves,
> to reassemble to make an airplane, or a boat, or a house.

It is not at all like that. It is like the fact that all three could
use a motor and all could borrow that motor from a car. All three could
use a seat, and all could borrow that seat from a car. All three could
use a windshield as a window. And complex cellular structures look like
they were tinkered together rather than intelligently designed.

> Don't you see, just because the subparts are functional as parts of
> different systems of function does not mean that these subparts can
> simply make an entirely new collective system of function. This just
> doesn't happen although evolutionists try and use this argument all
> the time. It just doesn't make sense. It is like throwing a bunch of
> words on the ground at random saying, "Well, they all work as parts of
> different sentences, so they should work together to make a new
> meaningful sentence." Really now, it just doesn't work like this.

What prevents the types of intermediate steps proposed for the bacterial
flagella? What prevents a TTSS from secreting and forming a whip-like
injectosome without motility function? What prevents such a system from
interacting with mot proteins that already function in a similar way
with other systems? What is to prevent a P-ring, used for generalized
export of materials, from being co-opted to a specialized function as a
bushing for flagella? Which of the several steps (we are talking about
the several steps that make this a chain of two-body events rather than
a magical poofing together of an n-body system from its parts) mentioned
as intermediate steps is 'unevolvable'?

> You must be able to add the genetic words together in a steppingstone
> sequence where each addition makes a beneficial change in the overall
> function of the evolving system. If each change does not result in a
> beneficial change in function, then nature will not and cannot select
> to keep that change. Such non-beneficial changes are either
> detrimental or neutral. The crossing of such detrimental/neutral gaps
> really starts to slow evolution down,

Yes to all the above. But that is indeed how evolution works.

> in an exponential fashion,
> beyond the lowest levels of specified functional complexity.

This last, however, is meaningless verbiage. You have not yet told
anyone how to determine what you mean by "lowest levels of specified
functional complexity". How do you determine the "level of specified
functional complexity"? What is the metric you use? Until you tell us
how you determine these numbers or how you are able to determine that it
is "thousands of amino acids" in bacterial flagella (for example) or
even how you determine it is 400 aa in lactases and why number of amino
acids says anything about *functional* complexity, you are engaged in
producing a pseudoscientific pseudocalculation which means little more
than saying "I find it difficult to imagine how this could have evolved
by magically poofing into existence from utterly functionless pieces of
junk." Of course, no evolutionary mechanism proposes that any system
'magically poofs into existence' from 'utterly functionless pieces of junk'.

> Very
> soon, evolution simply stalls out and cannot make any more
> improvements beyond the current level of complexity that it finds
> itself, this side of zillions of years of average time.
>

Until we understand what you mean by "level of complexity" such that
adding a modification of or to the pre-existing "level of complexity"
that results in a change of function cannot happen, your calculation is
nothing more than the old whore of imagining a 747 assembling by a
random process in a tornado.

> Sean
> www.naturalselection.0catch.com
>

howard hershey

unread,
Jan 21, 2004, 3:21:45 PM1/21/04
to

Sean Pitman wrote:

> jethro...@bigfoot.com (Jethro Gulner) wrote in message news:<edf04d4a.04011...@posting.google.com>...
>
>>I'm thinking TSS to flagellum is on the order of chocolate to
>>chocolate-fudge-brownie
>
>
> Now that's a serious stretch of the imagination. The TTSS system is a
> non-motile secretory system while the fully formed flagellar system is
> a motility system as well. The TTSS system requires 6 or so different
> protein parts, at minimum, for its formation while the motility
> function of the flagellar system requires and additional 14 or so
> different protein parts (for a total of over 20 parts) before its
> motility function can be realized.

Wherein Sean exhibits confusion of what modern eubacterial flagella does
include and what it must include in order for there to be a motility
function. The modern eubacterial flagella includes parts that are not
*necessary* for motility. One does not *need* a whole bunch of
different closely related whip proteins to have a motile whip. And
several non-motile 'whips' do exist in association with TTSS systems
that involve smaller numbers of proteins.

If a whip exists and it is rotated, motility will occur. Subsequent
duplications and divergence to produce a *better* whip does not affect
the crucial protein-protein interactions that allow a whip to
self-assemble. The real problem for generating *motility* is linking
the mot protein subsystem to the core of the TTSS-like structure. It is
not *even* necessary that the original selectable function of that
linkage be cellular motility. It could be involved in something as
prosaic as helping in the transport of whip proteins by the TTSS-like core.

> Unless you can find intermediate
> functions for the gap of more than a dozen required parts that
> separate the TTSS system from the Flagellar system, I'd say this gap
> is quite significant indeed, requiring at minimum several thousand
> fairly specified amino acids.

The independent motor subsystem from which the flagellar motor derived
undoubtedly, like the related non-flagellar ExbBD or TolQR systems, act
to produce rotary motion through a third protein via energy produced by
an ion channel. The motor subsystem in the flagella acts to produce
rotary motion through a third protein via energy produced by an ion
channel. [sarcasm on] Clearly this is a large functional gap to be
leaped. The flagellar motor has gone all the way from generating motion
through a third protein via energy from an ion channel in some
non-flagellar system to generating motion through a third protein via
energy from an ion channel in a possibly related system. Such a massive
change in function must require, at minimum, several thousand fairly
specified amino acids that differ from the non-flagellar motor to the
flagellar motor. [sarcasm off]

> Certainly this is not the same thing as
> roaming around the same island cluster with the same type of function.

It isn't? To go all the way from a motor that produces motion through a
third protein using an ion channel to generating a motor that produces
motion through a third protein using an ion channel requires a swim
across a vast gulf of function-space?

> The evolution form the TTSS island of function to the brand new type
> of motility function found in the flagellar island would have to cross
> a significant distance before the motility function of the flagellum
> could be realized.

It requires the modification of a single protein, and only to the extent
that that protein now links a pre-existing motor to a pre-existing TTSS
central core. That is, the additon of the motor to the TTSS may well
involve nothing more complicated than a change in a single binding site.
It certainly did not involve any major change in *function* in either
the TTSS-like subcomponent (it still acts as a protein transport device)
nor in the motor subcomponent (it still acts as a motor). Rather, the
two together generated a (at least potential) emergent function, motility.

> Such a distance could not be crossed via random
> walk alone this side of zillions of years in any population of
> bacteria on Earth.

See Sean wave his hands. Wave your hands, Sean. Wave them furiously.
Maybe somebody will ignore the man behind the curtain pulling the chains.

> In order for evolution to have truly crossed such
> a gap, without intelligent design helping it along, there would have
> to be a series of closely spaced beneficial functions/sequences
> between the TTSS and the motility function of the flagellum.

TTSS with injectosome + motor = crude motility of injectosome. [There
are other alternatives as well. Such as TTSS with motor = improved
transport. This plus increasing length of injectosome = crude
motility.] Motility is either a surprise function or an emergent one.
A similar event undoubtedly is involved in the motility of the archaeal
flagella, just with fewer proteins.


>
> Where is this series of steppingstones? That is the real question!

Did you read Nic's article?

> Many have tried to propose the existence of various stepping-stone
> functions, but none have been able to show that these steppingstones
> could actually work as no one has ever shown the crossing from any
> proposed steppingstone to any other in real life. If you think you
> know better how such a series could exist and actually work to
> eliminate this gap problem, please do share your evolutionary sequence
> with us.

Did you read Nic's article? If so, perhaps you can post the stepwise
stepping stones he does and show why any of those stepping stones
(choose only one -- your best shot, or show why certain events cannot
happen in such a stepwise fashion) between the proposed functionally
useful independent intermediate structures require changes of "thousands
of fairly specified amino acids" or is *impossible* on mathematical or
logical grounds? The entire chain of stepping stones may have involved
a lot of changes, but asking for the entire chain as if it were a single
poofing event would clearly be teleological thinking and much closer to
what creationists think happened. And I agree that such a single event
poofing of a flagella is highly unlikely. Going from nothing or random
sequences directly to the end flagella would assume that all the
proposed intermediate functional states have no utility but to serve as
a precursor to the teleologic goal.

Science does not work by requiring that every last event must be tested
experimentally or it is regarded as unlikely. If, for example, it has
been demonstrated that single mutational events can cause two proteins
to bind to one another or interact with one another in a new way leading
to new or modified function, and the proposed stepping stone involves a
very similar sort of naturalistic event under similar conditions, the
normal inference would be that such an event is possible rather than
unlikely. If you have a *reason* as to why such a similar event is
unlikely in this case or must involve thousands of changes in this case,
do share it. Waving your hands and asserting that such an event is
impossible won't do in the face of the fact that similar events
(mutations that affect protein-protein interactions) have been observed.
It may be interesting in a vaguely 'everything is interesting' way to
experimentally demonstrate that such an event *could* link a motor
subsystem to a different system to produce rotory motion in that system,
but it would hardly tell us anything new about how the proteins work.
But why would it have to be done with a flagellar system in particular?
>
> Sean
> www.naturalselection.0catch.com
>

Von Smith

unread,
Jan 22, 2004, 1:28:44 AM1/22/04
to
"Zachriel" <an...@zachriel.com> wrote in message news:<100j472...@corp.supernews.com>...


Another typical evolutionary "Just 'O'" story.

Sean Pitman

unread,
Jan 22, 2004, 1:23:34 PM1/22/04
to
"Zachriel" <an...@zachriel.com> wrote in message news:<100j472...@corp.supernews.com>...
> "Sean Pitman" <seanpi...@naturalselection.0catch.com> wrote in message
> news:80d0c26f.03120...@posting.google.com...
> >
> > So, what you
> > "start with" is quite important to determining what is and what is
> > not beneficial. Then, beyond this, say you start with a short
> > sequence, like a two or three-letter word that is defined or
> > recognized as beneficial by a much larger system of function, such as
> > a living cell or an English language system. Try evolving this short
> > word, one letter at a time, into a longer and longer word or phrase.
> > See how far you can go. Very quickly you will find yourself running
> > into walls of non-beneficial function.
>
> First you made this challenge. I responded with a word puzzle where,
> starting with the single letter word "O", and by only changing one letter at
> a time, and with concatenation, I constructed the phrase, "Beware a war of
> words, Sean Pitman, ere you err."
>
> We can trace the "etymology" of each word used in the poem. Some of the more
> difficult words to create include "light", "choose", "instead" and "simple".

You think that if you have the individual words available that you can
just stick them together without regard to their collective
*beneficial* meaning or the possibility of mistaken/misaligned
insertions? Look at your "rules" again and note something very
interesting:

Rules: Change only one letter at a time from any existing string. Can
concatenate any two strings. However, only one operation at a time.
All words, phrases and sentences must make sense in standard English.

Your main problem here is your mistaken notion that you can
"concatenate" or simply connect, in a meaningful way, two separate
strings of independently meaningful code into a united stretch of
meaningful code. This is a common misconception among evolutionists.
However, the likelihood that two independent sequences will be united
so as to form a new collective function that is also
meaningful/beneficial in a given situation is inversely proportional
to the size of the final product - in an exponential manner - due to
the huge numbers of ways that they could be connected in a meaningless
way. You must consider the odds that they will concatenate themselves
in a meaningful way vs. all the huge numbers of
meaningless/non-beneficial possibilities that also exist.

For example, say that I have two phrases that read, "I like ice cream"
and "Life often follows a rocky road". Now, a mutation to the second
phrase could clip out the words "rocky road" and insert them just
right into the first phrase so as to create a new phrase, which reads,
"I like rocky road ice cream". This is certainly possible, but you
must ask yourself how likely such a perfect snipping and insertion
will be? I mean, the insertion could have read, "I lirocky roadke ice
cream" or the clipping could have been messed up from the beginning
and read "ows a rocky roa" and have been inserted to read, "I li ows a
rocky roake ice cream". Now, instead of having a new collective
meaning, the insertion of the new sequence(s) simply destroyed the
meaning of the previous sequence. In fact, real life experiments are
often done to identify the correct insertion of a particular genetic
sequence not only by detecting the production of this gene production
but also via the detection of a loss of a previously functional gene.
What happens is that the new gene gets inserted into the middle of a
previously functional gene. Such a random insertion almost always
destroys the function of the previous gene. The fact of the matter is
that above the lowest levels of functional complexity there are simply
no examples of genes concatenating themselves or even working together
to form a new collective function that is actually novel as well as
beneficial.

I'm afraid that your little scenario, although it must have taken you
quite some time to think of and write out, is fundamentally flawed.
Just because various sequences and codes already exist in a genome
does not mean that these sequences will be able to "concatenate"
themselves so neatly in a meaningful way as you seem to think vs. the
vastly more likely non-meaningful possibilities. And, this problem
only gets worse at higher and higher levels of specified complexity.
That is why the "homology" arguments are weak. Just because all the
necessary parts for a new beneficial function exist within a genome as
subparts of many other systems of function does not mean that all
these parts will simply come together in a meaningful collective way
to form an entirely new function beyond the lowest levels of
functional complexity - even if it would be highly beneficial if it
happened. You see, in order to work in a new way the parts must be
brought together in a very specific way, as you have illustrated
nicely in your concatenating example. The likelihood that such a
specified mindless placement will be meaningfully achieved at anything
beyond the lowest levels of specified complexity gets exponentially
more and more remote as you move up the ladder of specified
informational complexity.

So, you simple cannot get from "O" to "Beware a war of words, Sean
Pitman, ere you err" without crossing significant gaps of neutral or
even detrimental meaning/function. The simple "pasting together" of
pre-established words and phrases doesn't help you because you have
pasted them, with the help of your intelligent mind, in "just the
right way" so that they will work, when, in real life, it is far far
more likely that they will get pasted together in non-meaningful ways.

Sean
www.naturalselection.0catch.com

Bennett Standeven

unread,
Jan 24, 2004, 8:45:09 PM1/24/04
to
seanpi...@naturalselection.0catch.com (Sean Pitman) wrote in message news:<80d0c26f.04012...@posting.google.com>...

You misunderstand what he means by concatenation. The only valid
concatenations of these two phrases are: "I like ice cream life often
follows a rocky road" and "Life often follows a rocky road I like ice
cream." The chance of getting the desired concatenation of two phrases
is thus always 1 in 2, irrespective of their length.

Zachriel

unread,
Apr 25, 2004, 1:56:35 PM4/25/04
to

"Sean Pitman" <seanpi...@naturalselection.0catch.com> wrote
> "Zachriel" <an...@zachriel.com>

> STATISTICALLY IMPOSSIBLE . . . ZILLIONS OF YEARS

> This is a great story [Sea of Beneficence], but sadly, it is statistically
> impossible. . . . The problem here is that there simply is not enough
> time this size of zillions of years to get the limited number of phrases
> to "bump together" enough times to make anything beyond the lowest
> levels of functional complexity without the input of a higher intelligence
> or pre-established information system. It just won't happen.
> Try it and see.

Other famous Sean Pitman posts using the technical term "zillions".
http://tinyurl.com/2aokz


---------------------------
The LIMIT of CATS and DOGS

The basic thrust in this calculation will be to set an upper-limit to the
number of possible mutations per the rules of our game. We will use some
simplifying assumptions, but suffice it to say that our estimate will be
many orders of magnitude larger than the actual number of possible
mutations.

* Note that the original rules are a subset of these extended rules.

-------
DOGS

Let C = number of character-symbols in the language, called "letters". For a
basic English alphabet, it's 26, but let's make it 100 to make the
arithmetic a little easier. That also allows us to include numerals, spaces
and other "fancy" symbols. In fact, for large populations, this will not
even be a significant factor. We could just as easily make it a thousand.

Consider a simple case, "dog". Let L = length("dog") = 3. (We will slightly
modify our definition of length later.)

POINT-MUTATIONS(P): We can do a point-mutation on any one of the three
letters, or we can add a letter to either end. As there are 100 possible
letters, this would be a total of (L+2)*C = 500 possible point-mutations.
(We can also delete any single letter, but we will count these along with
the snippets.)

* We count P = (L+2)*C possible point-mutations.


SNIPPETS(S): From the word "dog", we can snip three different one-letter
sections "d" "o" "g", two different two-letter sections "do" "og", and just
one three-letter section "dog". This forms a triangular series 1+2+. . . L =
L*(L-1)/2 < half of (L+1)^2. This is the number of possible snippets from a
string that might create a new string, a free-snippet. Also, when we snip
out the "o", we leave "dg", which if it were valid, might also enter the
general population. The number of such remainder strings is also < half of
(L+1)^2. So there is an upper limit of S = (L+1)^2 new strings created by
snipping.
When we calculate the number of possible insertions, we'll overcount
somewhat and also let S = (L+1)^2. Why quibble over details?

* We count S = (L+1)^2 possible snippets and remainders


INSERTIONS(I): We can insert each of these snippets (S), in four different
places in the word "dog"; before the "d", before the "o", before the "g", or
at the end of the word. So I = S*(L+1) = (L+1)^2 * (L+1) = (L+1)^3

* Note that insertion at the beginning of the word or at the end is the same
as a concatenation.

* Let's make another simplifying assumption. From now on, we will treat
the length of a word such that L = length("dog")+2 = length(" dog ") = 5.
This will increase our count somewhat, but we won't have to use L, L+1, or
L+2 in different parts of the calculation. Let's give our calculation a
little room to breathe.

* This makes I = S * L = L^3, a nice round figure. Gee whiz. Maybe Sean
Pitman is right, after all. That number does increase geometrically!


MUTATIONS(M): Consider a pond filled with a large multitude of the word
"dog" with mutations occurring randomly among the population. To consider a
single change to a single string in our population, we will consider every
possible mutation. Most such mutations will be non-viable, i.e. not valid in
the English language, e.g. "dxg". However, a few will be valid and can be
selected for beneficence, or meaningfulness. If the number of possible
mutations (M) is orders of magnitude larger than the population of "dog",
that is, if M is in the "zillions", then such a beneficial mutation will
probably never occur. For " dog ", M = 5*100 + 5^2 + 5^3 = 650 possible
mutations. This is clearly less than "zillions." Certainly, a reasonably
large population of "dog" could evolve by these rules into "dogs" or "dig"
or "cog" or "do".

* By our reckoning, M < P+S+I = C*L + L^2 + L^3

* Note that for large L, the point-mutations (P) and free-snippets (S) are
negligible and can be disregarded. More on this later.


-------------------
CATS and DOGS

Now consider two words, "cat" and "dog". Create a new string for
consideration (not meant to be an actual mutation, just an aid in
computation), with a space at the beginning and end of each word,
" cat dog " (consistent with our new definition of L).

We can count each of the point-mutations on the combined string and use this
to calculate the sum of the number for each of the strings separately. We
can also count the number of snippets. Of course, we will get a few snippets
which include parts of both individual words, so our count will be high.
Depending on the number of words and the length of the words, our count
might be way high. But that's ok. Why quibble over a few orders of magnitude
here or there?

L = 10
P = C*L = 100*10 = 10^3
S = L^2 = 10^2
I = L^3 = 10^3
M = P+S+I = 10^3 + 10^2 + 10^3 = 2100.

Now, Sean Pitman claims that anything over seven-letters has millions of
possible permutations. This is incorrect. An upper-limit for 20-letters
(which might be a bunch of small words, a few larger ones, or a mixture of
phrases and words) is 8,000. Most of these, as Sean Pitman correctly points
out, are not valid words or phrases, and can be automatically excluded from
the next generation.


-------------------
The MENAGERIE

Now, consider a collections of many words, with a total length of 1000
letters, including the extra spaces as separators.

L=1000
P = L*C = 1000*100 = 10^5
S = L^2 = 10^6
I = L^3 = 10^9

M = P + S + I = 10^5 + 10^6 + 10^9 = ~10^9

* Note that the point-mutations and free-snippets are negligible for
large numbers. Considering all our simplifying assumptions, for a large
population we can treat M = L^3. For a thousand letters, the total possible
mutations is 10^9, which is many orders of magnitude less than "zillions".

The verse "Beware a war of words, Sean Pitman, ere you err." was evolved in
a space of about 1000 letters, including many words that are not needed for
the final derivation. You can see this in "A Pond of Doggerel". We could
even optimize our process to fit in a smaller pond. More on this in
"Malthusian Catastrophe".

Now, for the entire poem, "O Sean Pitman", the total length of every phrase,
word, space and comma in the entire project is less than 5000. 5000^3 is
10^11, still much less than our pond-size of 10^14. You can find the
complete evolution at the beginning of the thread.

O Sean Pitman
http://tinyurl.com/2rw58

* Quo erat demonstratum.

Zachriel

unread,
Apr 25, 2004, 2:04:08 PM4/25/04
to


> STEPPINGSTONES

> Without these steppingstones, which are simply not close enough
> very far beyond the lowest levels of functional complexity, evolution
> stalls out completely this side of zillions of years.

Sean Pitman steps on "steppingstones".
http://tinyurl.com/39t9o

--------------------------
MALTHUSIAN CATASTROPHE

You may have seen this little puzzle earlier in the thread. Using our
original rules of point-mutation and concatenation, we can derive this line
of verse from Hamlet. (Each line in the puzzle is arbitrary. We are not
formally distinguishing each generation.)

--------------------------
o, or, to, no, not
i, bi, be

to be
or not
to be or not
to be or not to be
--------------------------

Counting every single word and phrase, the puzzle takes up only 50
characters, and yet exhibits intense meaning in the English language.

M = L^3 = 50^3 = 125,000

Malthus, in case you don't remember, wrote "An Essay on the Principle of
Population" that basically stated that a population will increase
geometrically until it exceeds its available resources. We have always
assumed that once a word is evolved, it remains in the population. But what
if the pond is limited in size? Perhaps less meaningful, words are culled or
go extinct. So can we use Malthusian Catastrophe to be even more
parsimonious in our population size? In other words, will extinction allow
our little game to occur in a much smaller pond? (This time, each line is
now a distinct generation along with its population. Many strings go extinct
and are not carried forward, including "o" and "or not".)


-----
o
o, i, a
or, to, no, bi
be, to, or, not
to be, or not
to be, to be or not
to be or not to be
-----

So, as you can see, we evolved the phrase "to be or not to be" in a space of
less than 20 letters, just enough for the final phrase, and did it in just 6
generations. M < L^3 = 8000.

Because of selection, because we limit our population in each generation by
*whatever rule* we have decided upon, the total choices that must be
considered over many generations are on the order of M1+M2+M3 . . . = M*G,
and not M^G. See "Elementary Arithmetic, My Dear Pitman" for more on this.


---------------------------
MASS EXTINCTION!!

Now for fun, here is the evolution of the next phrase in Hamlet's immortal
soliloquy. We again start with the single-letter word, "O", the last
surviving word from our previous population (after a recent lexiconic
catastrophe).

o, i, is, his, him, hem, he, the
a, at, hat, that
on, ion,
id, qid, quid, quiz, quit, quiet, quiets,
eat, seat, set, let, lest
gest, guest, quest
question

that is
the question
that is the question

-----

The reader can easily see where extinction of unneeded words and phrases
could reduce the size of the puzzle. For instance, once we have the word
"the", we no longer need "his", "him", "hem", "he". Some words, like "quiz",
aren't needed at all. Think of them as evolutionary byways. Even without
optimization, the puzzle is not much more than 100 letters.


-------------------------
OUR P's and Q's

The "q"-word was rather tricky, but I found two ways to evolve it, one from
"guest", the other through "qid". But as with biological evolution, once we
have a "qu", we can use it all sorts of way.

quit, quilt, quill, quint, squint, squinty
quip, equip
quid, squid, equid, equip
quad, squad, squid
quite, quote, quire, quirk, quark
squire, squirm, esquire

And this is without considering any of Pitman's Patented Slicing-and-Dicing,
but only point-mutations and concatenation. In language-arts, these kinds of
word-games can be very helpful. Consider that we might use the word "equid"
rather than "horse" just because of the happenstance of word-evolution or
mental association. Biological evolution exhibits an analogous process of
opportunism.

--------------------
A ZILLION ROADS DIVERGED

So at an intersection of roads, there are, let's say, just two possible
directions we can go, and each of these roads diverge into another two
others, and each of these roads continue to diverge a hundred times over. To
explore each and every possibility would mean taking every possible choice,
and every possible road, at every single intersection. Counting just one
mile of walking between each intersection (and not including the distance we
must travel to retrace our steps as we back up to each and every node), we
would have to travel 2^100 = 10^30 miles. But how far would we have to walk
to reach the end of just one possible journey at the far end of the road
system? Just a hundred miles, a hundred choices, and a week's journey.

We may have a zillion choices, but we choose just one. Whether the selection
is that of a breeder, a traveler, or of natural selection, the choice is
made. Some travelers may choose the low road, others the high road. Some
looking for company may choose the well-traveled road, others looking for
adventure may choose the lonely road. And some may peer ahead as far as they
can, and see which has the greener grass. We make a choice. We discard the
rest. And all those zillions of other possibilities are left behind. And to
the traveler, this makes all the difference.


TWO roads diverged in a yellow wood,
And sorry I could not travel both
And be one traveler, long I stood
And looked down one as far as I could
To where it bent in the undergrowth;

Then took the other, as just as fair,
And having perhaps the better claim,
Because it was grassy and wanted wear;
Though as for that the passing there
Had worn them really about the same,

And both that morning equally lay
In leaves no step had trodden black.
Oh, I kept the first for another day!
Yet knowing how way leads on to way,
I doubted if I should ever come back.

I shall be telling this with a sigh
Somewhere ages and ages hence:
Two roads diverged in a wood, and I—
I took the one less traveled by,
And that has made all the difference.

Robert Frost
http://www.bartleby.com/people/Frost-Ro.html


Zachriel

unread,
Apr 25, 2004, 2:01:58 PM4/25/04
to

"Sean Pitman" <seanpi...@naturalselection.0catch.com> wrote
> "Zachriel" <an...@zachriel.com>

> YOU SIMPLY CANNOT

> So, you simple cannot get from "O" to "Beware a war of words, Sean
> Pitman, ere you err" without crossing significant gaps of neutral or
> even detrimental meaning/function.

Sean Pitman discusses his "significant gaps".
http://tinyurl.com/2vv73

-------------------
A POND of DOGGEREL

Using our original rules of point-mutation and concatenation, we will evolve
a simple line of verse. The original rules are a subset of the extended
rules.

-------

o, a, i
o, or, ore, one, wore, word, whore

words, wordy, ward, war, tar, wars, ware, tare, are
ere, err, era, ore, ode, of, off, or, our, your, you
ire, irk, irks, lire, lyre, fire, lice, like, lick,
lock, block, click, clock, slick, stick, stack
for, fore, form, forms, foreword
ow, row, brow, prow, prom, from

war, wan, man, may, mean, many, bean, bear, beer, bee
be, ear, year, dear, tear, pear, spear, dean, deal, ideal, idea

Sean, sear, bead, lead, seer, steer, steed, stead


eat, ear, seat, set, wet, we, see, sit, pit, it, is, in, gin, instead
seep, step, pet, poet, poem

ion, sir, stir, stair, staid, tee, tea, teat, tear
treat, great
as, an, can, and, ass, pass, piss, kiss
to, do, so, go, no, not, nod
sin, tin, kin, king, win, wine, pine, pin, ping
wee, weeping, weening, weaning, meaning
is, his, this, him, hem, he, the, thy, why, who, thin, think, thing

-----

be-ware


wordy ward, a wordy ward, word wars
a war, a kiss, world, world war

of words, a war of words
beware a war of words
pit-man, Sean pitman
beware a war of words Sean pitman
you err
I err
ere you err

a war, of words, a war of words
beware a war of words
pit-man
Sean pitman
beware a war of words Sean pitman
you err
I err
ere you err

* Beware a war of words, Sean Pitman,
* Ere you err.


-----
The BLIND PIG and the ACORN

In a space of a thousand letters, we have evolved a phrase with significant
meaning. In addition, we have evolved a large of words that can be used to
evolve many other phrases with significantly different meanings, e.g "a bee
and a bear".

There are two types of selection criteria involved in our discussion;
selection for phrases with *specific* meaning, that is, goals analogous to
animal husbandry; and undirected evolution for phrases with *any* meaning
whatsoever. So, if we want to find a phrase with a specific meaning, we
select from our available mutations words that are hopeful. For instance, if
we want to create a phrase meaning "contemplating suicide", we might select
the word "be", which means to exist. On the other hand, if we don't care
about specific meaning, but any meaning will do, then evolution can meander
about. We might select "a bee and a bear", rather than "a steer and a lyre",
or visa versa; while tossing out other viable candidates. The important
point is that selection occurs to narrow our future options.

Sean Pitman claims, "The lower the density, the harder it is to evolve new
stuff." This is somewhat misleading. We can assume he means that the
connections between valid phrases, instead of being wide avenues, are
actually thin tendrils, and that these tendrils become stretched as we
increase the number of letters, until they finally snap and there is no way
to get from "a blind pig" to "the acorn". However, we have shown that such a
path exists all the way from "O" to "O Sean Pitman". So in fact, if a
particular path exists, then the fewer the choices, the *easier* it is to
find the specific path.

You can find the entire poem "O Sean Pitman", and the story of the Sea of
Beneficence, earlier in the thread, here:

O Sean Pitman
http://tinyurl.com/3254c

At each generation, selections must be made for evolution to work. The
choices must be reduced to a few after each generation. The key is the
selection criteria, but as Sean Pitman has repeatedly pointed out, the vast
majority of mutations are not even valid words or phrases, claiming a ratio
for seven-letter words of 250,000 junk sequences for each meaningful one.
This reduces the problem by many orders of magnitude. We can select
artificially, like the horticulturalist, or we can allow the process to
occur based upon some rule of meaningfulness, such as assigning higher
meaning-quotients to more complex phrases.

In any case, as long as a path exists, evolution can continue. So if we
don't care about finding a phrase with a specific meaning, then evolution
can meander about, finding longer and longer phrases, just as meaningful in
their own way as "O Sean Pitman". For example, from our existing population
of words, we can easily see the evolution of a phrase such as this one,

A bee and a bear kiss by a fire
While Sean Pitman plays the lyre.

or a multitude of other much like it. It is important to note that our
evolutionary process won't find every possible combination of words, or even
any specific combination of words. But it will find some of them.

-----
WHAT CAN I SAY

Well, we have not actually shown that we can evolve ANY possible string of
text. Some words and phrases may be out-of-reach of our methods. With the
original rules, there do appear to be problems finding routes to specific
words, but there always seems to be another word or phrase that will do just
as well. On the other hand, we do know that the poem "O Sean Pitman" can be
evolved, and that clearly indicates the power of mutation and selection in
regards to our game.

The biological Theory of Evolution does not claim that every imaginable
creature can be evolved, and certainly not that every imaginable creature
must evolve and co-exist. Rather, like our game, evolution is opportunistic;
and though every conceivable creature may not be evolvable, the diversity of
life on Earth clearly indicates that the range of evolvable organic forms is
vast.


-----
ZILLIONS OF SEAN PITMAN'S

Perhaps we can't evolve, for instance, from "painter" to "eternal", but does
that necessarily mean that we can't evolve from "Sean Pitman" to "just plain
wrong"?

According to Sean Pitman's calculation, we would have to consider each and
every possible combination of letters and spaces for the 16 letters that
make up "just plain wrong", 26^16 or 10^22, a very large number indeed. But
we don't have to look at every single combination, but only at those that
are available in each generation. In addition, we can immediately discard
invalid words and phrases, and need consider only the ones that appear to
get us closer to our goal (which for a breeder might be a slightly stronger
horse, or a dog with its nose a bit closer to the ground).

Sean Pitman
pitman
pit, man
put, pan, wit
jut, plan, wig
just, plain, wing
just plain, wring
just plain, wrong
just plain wrong

Let L = total length of all extant string species
Let M = total possible mutations
Let G = number of generations

We have shown that the number of possible mutations under the extended rules
is M < L^3 per generation. Including spaces and commas, the largest L is 17
in length, so M < 5000 for each generation (most of which are not even valid
words or phrases). The selection criterion is obvious in this case; pit,
put, jut, just. The total choices we must make are not 5000^G as Sean Pitman
would suggest, but on the order of 5000*G. It is the process of selecting
that makes it a product rather than an exponent. (And the choosing is
obvious in this case.)

In a Sea swimming with trillions of Sean Pitman's, some just might evolve
into new and different forms.


Sean Pitman

unread,
Apr 30, 2004, 12:31:26 AM4/30/04
to
"Zachriel" <sp...@zachriel.com> wrote in message news:<1cmdnUj5jq5...@adelphia.com>...

Since you are quoting me on your website Zach, it might be good to
note that I never said that a 7-letter sequences would take "zillions"
of generations much less years to evolve. In fact, I have said just
the opposite many times. What I said was that 7-letter sequences
would be exponentially more difficult to evolve, on average, than
those requiring fewer letters, like 3-letter sequences. However, if
you increase the population size (given a constant mutation rate and
generation time) in an equivalently exponential manner as sequence
size increases, meaningful evolution will proceed with the same ease
as it did at the lower levels of minimum sequence size.

Now, consider again your suggested population of 10^14 (100 trillion).
With such a population you could cover all of the sequence space for
7-letter sequences 12,000 times over. Obviously then you are grossly
misstating the strength of your argument. The evolution of not just
every meaningful word, but every _possible_ sequence in 7-letter
sequence space could be achieved by your rather large population in
very short order (i.e., less than one generation at population
equilibrium within sequence space).

I am also well aware that language systems, such as the English
language system and even biological/genetic language systems tend to
cluster meaningful symbols together. I am also aware that as you
increase the minimum sequence length, these clusters start to separate
from each other and become relatively smaller at the same time so that
it becomes exponentially more and more difficult to evolve from one
cluster to the next without having to cross through oceans of
non-beneficial sequences.

So, given this position of mine, why don't you try to set up your
computer to evolve using the following rules, if you agree to these
rules, and see if they don't match my predictions better than yours:

Genome size:
1,000 characters

Population size:
100 trillion (10^14) genomes existing in a steady state over the
course of all generations

Mutation rate:
1 mutation per genome per generation

Types of mutations:
Point, deletion, insertion, and recombination (note that all
deletions, insertions, and recombination mutations must be random with
respect to the number of characters involved as well as the position
of the mutation in the genome)

Valid characters:
26 English letters, period, and space (28 total)

Beneficial selection:
Any string that has any portion of itself making meaningful sense in
the English language system will be accepted as "beneficial" in this
experiment. Those strings that have longer portions making meaningful
sense will be given a greater survival advantage ranging from 1 to
1,000 advantage points.

Starting point:
Each one of the 100 trillion genomes starts out with different
meaningful 7-character sequences (words or meaningful 7-character
phrases) repeated over and over again for a total of 1,000 characters.
Note that you will be able to cover not only all meaningful
7-character sequences in the English language system, but all possible
7-character sequences in 7-character sequence space thousands of times
over.

Goal:
Calculate how many generations it takes to evolve just 1,000
meaningful sequences at each level of complexity ranging from 1 to
1,000 characters. Obviously you will most likely start out covering
everything meaningful ranging from1 to 7 characters in size right at
your starting point. So, you can list all of these levels as taking
"zero generations" for your population to achieve success.

It seems very likely to me that the next higher levels (i.e: 8, 9, 10,
etc) will take only one or two generations for your population to
evolve 1,000 uniquely meaningful sequences at each level. However, by
the time you get to level 25, I am thinking that your population is
going to start noticeably stalling in its ability to evolve the 1,000
uniquely meaningful English sequences. By level 50 I'm not sure that
your population of even 100 trillion will succeed in less than a
million generations. And, I feel fairly confident that by level 100
your population will do exceptionally well to succeed this side of a
trillion generations.

Certainly you can see where I am going with this, but I'll do it
anyway. I predict that the generations needed to get to level 1,000
in this little game of evolution would be well over 10^1000 - truly
"zillions" of generations indeed!

Now, see if you can prove that my predictions here are significantly
in error . . .

Good luck with your "war of words." ; )

Sean
www.naturalselection.0catch.com

Andrew Arensburger

unread,
Apr 30, 2004, 1:07:08 PM4/30/04
to
In talk.origins Sean Pitman <seanpi...@naturalselection.0catch.com> wrote:
> So, given this position of mine, why don't you try to set up your
> computer to evolve using the following rules, if you agree to these
> rules, and see if they don't match my predictions better than yours:

> Genome size:
> 1,000 characters

> Population size:
> 100 trillion (10^14) genomes existing in a steady state over the
> course of all generations

Just storing one generation would require 10^17 characters. I
make that out to be nearly 89 petabytes. I doubt that Zach has that
kind of storage available.
It might be possible to compress this data, but unless you can
suggest a scheme for achieving a compression ratio of 930000, I don't
think this entire population will be able to fit on an off-the-shelf
disk for an average desktop PC.

Other than that, is 10^14 doesn't seem like a reasonable
population size. Can you name some species with 10^14 members?

> Mutation rate:
> 1 mutation per genome per generation

Now there's the problem of time: assuming that we can generate
3 billion mutations per second (a number obtained by starting with my
2.5 GHz processor, rounding up to 3 GHz, and making the obviously
unreasonable assumption that we can generate one mutation per
instruction), it would take close to six years just to mutate one
generation.
So if you're serious, you're going to have to relax your
requirements.

> Types of mutations:
> Point, deletion, insertion, and recombination (note that all
> deletions, insertions, and recombination mutations must be random with
> respect to the number of characters involved as well as the position
> of the mutation in the genome)

> Valid characters:
> 26 English letters, period, and space (28 total)

> Beneficial selection:
> Any string that has any portion of itself making meaningful sense in
> the English language system will be accepted as "beneficial" in this
> experiment. Those strings that have longer portions making meaningful
> sense will be given a greater survival advantage ranging from 1 to
> 1,000 advantage points.

Please elaborate. Be specific enough that this can be coded.
For one thing, "meaningful" is not an all-or-nothing quality,
but comes in degrees. For instance, when a friend of mine told me
"ek.", I knew that something was the matter, but didn't know what
exactly. With just one mutation, we could get to "ok.", which is,
intuitively, more meaningful. But how much more meaningful? How do you
quantify this?
Secondly, a string consisting of randomly-chosen English words
is word salad, but is closer to being meaningful than a string
consisting of randomly-chosen letters. A real English sentence, but
with a typo in every other word, is even closer to being meaningful,
even if it contains a high number of words that aren't proper English
words. Again, how do you quantify this? (Aside: is "Finnegans Wake"
meaningful?)
Thirdly, your criterion above counts string length, but fails
to take into account the quality of the meaningful string. By this
criterion, "ok. ok. ok. ok. [... repeat to 1000 characters...]" or
"hello. hello. hello. [... repeat to 1000 characters...]" will have a
higher score than "shall i compare thee to a summers day.
ttagggttagggttagggttaggg [... repeat "ttaggg" to 1000 characters...]".

> Starting point:
> Each one of the 100 trillion genomes starts out with different
> meaningful 7-character sequences (words or meaningful 7-character
> phrases) repeated over and over again for a total of 1,000 characters.
> Note that you will be able to cover not only all meaningful
> 7-character sequences in the English language system, but all possible
> 7-character sequences in 7-character sequence space thousands of times
> over.

As I said above, it is unreasonable to even _store_ the
starting population. How about just storing a subset? A few hundred or
thousand sequences?

> Goal:
> Calculate how many generations it takes to evolve just 1,000
> meaningful sequences at each level of complexity ranging from 1 to
> 1,000 characters.

What do you mean by "level of complexity"? Do you mean "length
of meaningful substring"?

> Obviously you will most likely start out covering
> everything meaningful ranging from1 to 7 characters in size right at
> your starting point. So, you can list all of these levels as taking
> "zero generations" for your population to achieve success.

> It seems very likely to me that the next higher levels (i.e: 8, 9, 10,
> etc) will take only one or two generations for your population to
> evolve 1,000 uniquely meaningful sequences at each level.

You say "it seems," which is just a statement of your
intuitive notion. How do you plan to test whether your intuition is
correct?

> However, by
> the time you get to level 25, I am thinking that your population is
> going to start noticeably stalling in its ability to evolve the 1,000
> uniquely meaningful English sequences.

Again, how do you plan to test whether what you are thinking
is correct?

> By level 50 I'm not sure that
> your population of even 100 trillion will succeed in less than a
> million generations.

Again, how do you plan to test this doubt of yours?

> And, I feel fairly confident that by level 100
> your population will do exceptionally well to succeed this side of a
> trillion generations.

Again, how do you plan to find out?

> Certainly you can see where I am going with this, but I'll do it
> anyway. I predict that the generations needed to get to level 1,000
> in this little game of evolution would be well over 10^1000 - truly
> "zillions" of generations indeed!

You seem to think that it's harder to start with a bunch of
meaningful 24-letter sequences and find a meaningful 25-letter
sequence, than it is to start with a bunch of meaningful 7-letter
sequences and find a meaningful 8-letter sequence. I don't understand
why you think that. Why don't you show your math?

In fact, since your mutation rules allow recombination, then
it's quite possible to make comparatively large leaps in length: if
"xyzzyxyzzyxyzzy hello ccgattccgatt[...] (meaningful length 5-7,
depending on whether you count the spaces) merges with
"ss.yxss.yxss.yx.ssnxy world uuspxm[...]" (meaningful length 5-7,
again), you can wind up with "xyzzyxyzzyxyzzy hello world uuspxm[...]"
(meaningful length 11-13).
So one generation can effectively double the length of the
longest meaningful sequence.

> Now, see if you can prove that my predictions here are significantly
> in error . . .

Sure, but what you've suggested so far is not implementable.
It seems to me that using a smaller generation size would make
the program implementable. If it succeeds in, say, 8106 generations
with a generation size of, say, 500, then that would place an upper
bound on what could be achieved, in principle, with a generation size
of 10^14.
There remains the crucial question of how to determine whether
one candidate is more fit than another. It is difficult for a program
to determine whether a given sequence is grammatically correct or not,
meaningful or not. Or, which is even more interesting, _close_ to
being grammatically correct or meaningful.
We could simplify this by seeing whether the program can
evolve a particular 1000-letter sequence (or one of a small set of
1000-letter sequences). Since the program would be trying to hit a
much smaller target than the set of all possible meaningful
1000-letter sequences, this would place an upper bound on the
performance of the full program.
However, this now looks an awful lot like Dawkins's "Weasel"
program, which we already know doesn't take zillions of generations to
run.

--
Andrew Arensburger, Systems guy University of Maryland
arensb.no-...@umd.edu Office of Information Technology
On second thought, let us not go to Z'ha'dum. It is a silly place.

Zachriel

unread,
May 1, 2004, 8:56:14 AM5/1/04
to

"Sean Pitman" <seanpi...@naturalselection.0catch.com> wrote in message
news:80d0c26f.04042...@posting.google.com...

> "Zachriel" <sp...@zachriel.com> wrote in message
news:<1cmdnUj5jq5...@adelphia.com>...

<snipped>


> Beneficial selection:
> Any string that has any portion of itself making meaningful sense in
> the English language system will be accepted as "beneficial" in this
> experiment.

Um, how do you rigorously define "meaning"?

Are these meaningful phrases?

"dog"
"a dog"
"the cat barks"
"I drown an eye"
"soap, bread, milk, jalapeno peppers"
"milk, flour, sugar, vanilla, baking powder"

Does anyone have a meaning-o-meter I can borrow? It has to work fast,
though, about a millisecond would be fine.

Zachriel

unread,
May 1, 2004, 8:55:57 AM5/1/04
to

"Sean Pitman" <seanpi...@naturalselection.0catch.com> wrote in message
news:80d0c26f.04042...@posting.google.com...

> "Zachriel" <sp...@zachriel.com> wrote in message
news:<1cmdnUj5jq5...@adelphia.com>...
>
> Since you are quoting me on your website Zach, it might be good to
> note that I never said that a 7-letter sequences would take "zillions"
> of generations much less years to evolve. In fact, I have said just
> the opposite many times. What I said was that 7-letter sequences
> would be exponentially more difficult to evolve, on average, than
> those requiring fewer letters, like 3-letter sequences. However, if
> you increase the population size (given a constant mutation rate and
> generation time) in an equivalently exponential manner as sequence
> size increases, meaningful evolution will proceed with the same ease
> as it did at the lower levels of minimum sequence size.
>
> Now, consider again your suggested population of 10^14 (100 trillion).
> With such a population you could cover all of the sequence space for
> 7-letter sequences 12,000 times over. Obviously then you are grossly
> misstating the strength of your argument.

Um. In case you didn't know, I didn't directly simulate a population of 100
trillion. My genomes had a population size of just one, yes that's
right, just one for each species, the number of species limited to the few
best (default 25), and their mutants which have at most one mutation each.


> The evolution of not just
> every meaningful word, but every _possible_ sequence in 7-letter
> sequence space could be achieved by your rather large population in
> very short order (i.e., less than one generation at population
> equilibrium within sequence space).

It can be achieved by a very small population in a split second.


> I am also well aware that language systems, such as the English
> language system and even biological/genetic language systems tend to
> cluster meaningful symbols together. I am also aware that as you
> increase the minimum sequence length, these clusters start to separate
> from each other and become relatively smaller at the same time so that
> it becomes exponentially more and more difficult to evolve from one
> cluster to the next without having to cross through oceans of
> non-beneficial sequences.

You have not shown that. Indeed, there is no way to know that from mere
mathematical analysis. You have to know their distribution. For all we know,
words are all lined up nice and pretty in "permutation space". It turns out
that many, perhaps most, of them are!


> So, given this position of mine, why don't you try to set up your
> computer to evolve using the following rules, if you agree to these
> rules, and see if they don't match my predictions better than yours:
>
> Genome size:
> 1,000 characters

Genome size should vary, of course, just like it does in the current
Mutagenator.


> Population size:
> 100 trillion (10^14) genomes existing in a steady state over the
> course of all generations

You may own a computer, but you apparently don't know much about them. I
don't have 10^14 bytes of RAM in mine. Do you?


> Mutation rate:
> 1 mutation per genome per generation

That's an interesting limitation in a population of trillions. If we were to
actually model a large population, then mutations would be occurring
simultaneously throughout the population. A better solution would be to just
keep one copy of each species in memory, pick a string randomly, then mutate
it; or if recombining, then pick two strings randomly, take a random snippet
from the first and insert it in a random position in the second. The point
is that we don't have to model trillions of copies to test random mutation.
Somewhere a random portion of a random string is snipped and then reinserted
in another random string. If it is "superior", it is kept. If not, it is
discarded.

This is exactly how the Word Mutagenator works.


> Types of mutations:
> Point, deletion, insertion, and recombination (note that all
> deletions, insertions, and recombination mutations must be random with
> respect to the number of characters involved as well as the position
> of the mutation in the genome)

This is exactly how the Word Mutagenator works.


> Valid characters:
> 26 English letters, period, and space (28 total)
>
> Beneficial selection:
> Any string that has any portion of itself making meaningful sense in
> the English language system will be accepted as "beneficial" in this
> experiment. Those strings that have longer portions making meaningful
> sense will be given a greater survival advantage ranging from 1 to
> 1,000 advantage points.

Any string or any portion? This may not be necessary. If I bother, I'll
probably require the entire string to make a complete thought in some sort
of "pigeon" English.


> Starting point:
> Each one of the 100 trillion genomes starts out with different

If I endeavor to expand this project, I'll probably build something like the
Mutagenator which has a population of exactly one for each genome, then
mutates and recombines them randomly.


> meaningful 7-character sequences (words or meaningful 7-character
> phrases) repeated over and over again for a total of 1,000 characters.

What's this all about? Why not start with "O"? Or let people pick their own
words or phrases? Gotta make it a little interactive, you know.


> Note that you will be able to cover not only all meaningful
> 7-character sequences in the English language system, but all possible
> 7-character sequences in 7-character sequence space thousands of times
> over.
>
> Goal:
> Calculate how many generations it takes to evolve just 1,000
> meaningful sequences at each level of complexity ranging from 1 to
> 1,000 characters.

I don't think you mean calculate, but generate. If it is done randomly, then
each run will be unique. Indeed, your requirement here makes little sense.

How 'bout we just generate phrases on the order of "to be or not to be", or
"beware a war of words ere you err". That should prove the point
sufficiently.

Or have we pushed your incredulity to such an extent that to push it further
we have to directly model all 100 trillion genomes of 1000 bytes each for
millions of generations?


> Obviously you will most likely start out covering
> everything meaningful ranging from1 to 7 characters in size right at
> your starting point. So, you can list all of these levels as taking
> "zero generations" for your population to achieve success.
>
> It seems very likely to me

Show your math.


> that the next higher levels (i.e: 8, 9, 10,
> etc) will take only one or two generations for your population to
> evolve 1,000 uniquely meaningful sequences at each level.

Now you want 1000 sequences of 1000 each? That's a megabyte. How about we
make one, and then run the problem a thousand times.


> However, by
> the time you get to level 25, I am thinking that your population is
> going to start noticeably stalling in its ability to evolve the 1,000
> uniquely meaningful English sequences. By level 50 I'm not sure that
> your population of even 100 trillion will succeed in less than a
> million generations.

You keep going on about this 100 trillion. Well, there may be that many
bacteria in your gut, but there isn't that much memory in your computer.
Please return to common-sense. We need to do this in thousands to make the
computation feasible with the tools we have available.

Keep in mind, this is your word-analogy. It was you who would not heed the
warning to "beware a war of words." You said to start with short words and
try to evolve longer words and that we would reach "walls" beyond which we
couldn't go. This assertion has already been falsified, though you refuse to
admit it.


> And, I feel fairly confident that by level 100
> your population will do exceptionally well to succeed this side of a
> trillion generations.
>
> Certainly you can see where I am going with this, but I'll do it
> anyway. I predict that the generations needed to get to level 1,000
> in this little game of evolution would be well over 10^1000 - truly
> "zillions" of generations indeed!
>
> Now, see if you can prove that my predictions here are significantly
> in error . . .
>
> Good luck with your "war of words." ; )

Round one to Zachriel.

Round two may take a while. I'll let you know if I have time to rewrite the
program in a reasonable amount of time. The last project took a few weeks of
spare time. We'll have to take it to the steering committee once the rules
are hashed out.


Zachriel

unread,
May 1, 2004, 12:58:52 PM5/1/04
to
The latest version of Word Mutagenation is now uploaded.
http://www.zachriel.com/mutagenation/

Program Notes:

Only a few minor changes (and hopefully bugfree). I added a worksheet,
"Variables," to store the form settings. This should allow you to save your
settings along with the population between sessions. On the same worksheet,
you can change the time-out function for the word Mutagenator. This should
allow larger populations without having to repeatedly hitting the Mutagenate
Continuous button. To set the timeout to one year (forever), just set it to
0.

Try not to switch screens during the runtime of Word Mutagenation. It ties
up nearly all the system resources. If you insist, then when in your other
program and wanting to switch back, click on the Excel icon, and press Q.
That should stop the process and switch the screen. Also, you can press ESC
during runtime to interrupt the program flow and dump into Debug mode; or
when not in runtime, press alt-F11 to examine the source code.

Word Mutagenation has been scanned by Norton's Anti-Virus.

Sean Pitman

unread,
May 1, 2004, 9:02:22 PM5/1/04
to
"Zachriel" <sp...@zachriel.com> wrote in message news:<voKdnSHbCpP...@adelphia.com>...

> > Now, consider again your suggested population of 10^14 (100 trillion).
> > With such a population you could cover all of the sequence space for
> > 7-letter sequences 12,000 times over. Obviously then you are grossly
> > misstating the strength of your argument.
>
> Um. In case you didn't know, I didn't directly simulate a population of 100
> trillion. My genomes had a population size of just one, yes that's
> right, just one for each species, the number of species limited to the few
> best (default 25), and their mutants which have at most one mutation each.

A population of "one for each species" is not a population of one, but
many with "one for each species". Another thing, it seems like your
program analyzed thousands of sequences before it chose the 25 that
were "best".

If it were me, this is how I would set up the program: You start with
a maximum steady state population limit as well as a genome size
limit. In each generation you determine how many mutations, on
average, each one of the members of your population can sustain. Now,
the number of individuals in your population combined with the average
mutation rate per individual gives you the maximum number of sequences
that your population is allowed to analyze in a given generation. It
didn't seem to me that you did this with your program. But, please do
correct me if I am mistaken here.

> > The evolution of not just
> > every meaningful word, but every _possible_ sequence in 7-letter
> > sequence space could be achieved by your rather large population in
> > very short order (i.e., less than one generation at population
> > equilibrium within sequence space).
>
> It can be achieved by a very small population in a split second.

Although this is a small point, it would be more meaningful to talk in
generations here since modern computers can simulation huge numbers of
generations, depending on population size, in a "split second". In
any case, I would agree that 7-letter sequence evolution of quite a
few meaningful 7-letter sequences can be achieved quite rapidly with a
much smaller population than 100 trillion. I was just trying to point
out to you what a population of 100 trillion could achieve at this
relatively low level of meaningful sequence complexity.



> > I am also well aware that language systems, such as the English
> > language system and even biological/genetic language systems tend to
> > cluster meaningful symbols together. I am also aware that as you
> > increase the minimum sequence length, these clusters start to separate
> > from each other and become relatively smaller at the same time so that
> > it becomes exponentially more and more difficult to evolve from one
> > cluster to the next without having to cross through oceans of
> > non-beneficial sequences.
>
> You have not shown that. Indeed, there is no way to know that from mere
> mathematical analysis. You have to know their distribution. For all we know,
> words are all lined up nice and pretty in "permutation space". It turns out
> that many, perhaps most, of them are!

Actually, you have not shown that either. I say that the odds are
very strongly against that assertion. It is my position that all
language systems, to include English as well as genetic and protein
language systems of living cells are not lined up nice and pretty like
at all and that the clustering that does indeed exist at lower levels
of complexity get smaller and smaller and more and more widely spaces,
in and exponential manner, as one moves up the ladder of functional
complexity. This assertion is not only mathematically valid, it has
also been experimentally supported by many thousands of experiments
that have never show anything to evolve in any language system beyond
the lowest levels of functional complexity. For example, there are no
examples of protein functions evolving that require a minimum more
than a few hundred fairly specified amino acids working together at
the same time. And, this is despite well over 10^14 individual
organisms working on this problem under close observation for millions
of generations.



> > So, given this position of mine, why don't you try to set up your
> > computer to evolve using the following rules, if you agree to these
> > rules, and see if they don't match my predictions better than yours:
> >
> > Genome size:
> > 1,000 characters
>
> Genome size should vary, of course, just like it does in the current
> Mutagenator.

That's fine with me as long as you set an upper limit to genome size
(whatever limit you want as long as you stick to it).



> > Population size:
> > 100 trillion (10^14) genomes existing in a steady state over the
> > course of all generations
>
> You may own a computer, but you apparently don't know much about them. I
> don't have 10^14 bytes of RAM in mine. Do you?

This was just to make a point. Even if you could keep track of 10^14
genomes, it still wouldn't help you in this problem. But, use
whatever population size that your computer can keep track of - it
really doesn't matter as long as the population is limited to a steady
state at a certain point.



> > Mutation rate:
> > 1 mutation per genome per generation
>
> That's an interesting limitation in a population of trillions. If we were to
> actually model a large population, then mutations would be occurring
> simultaneously throughout the population.

What do you think a genome is? The population gene pool is made up of
all the individual genomes. For example, if you did have 100 trillion
individuals in your population, you would also have 100 trillion
genomes in your population with one genome per individual. That means
that with one mutation per genome on average, you would get 100
trillion mutations in your population in one generation.

> A better solution would be to just
> keep one copy of each species in memory, pick a string randomly, then mutate
> it; or if recombining, then pick two strings randomly, take a random snippet
> from the first and insert it in a random position in the second. The point
> is that we don't have to model trillions of copies to test random mutation.
> Somewhere a random portion of a random string is snipped and then reinserted
> in another random string. If it is "superior", it is kept. If not, it is
> discarded.
>
> This is exactly how the Word Mutagenator works.

I know that is how you programmed your "Word Mutagenator" to work, but
this is not a realistic enough comparison to real life evolution.

In real life, with the limit of a steady state population and genome
size, once you evolve a new and "better" sequence, you loose the
previous sequence. This is a problem, because it creates a rather
constant limitation on the pool of options that each generation can
use.

Also, in real life, genomes are not kept "in memory" immune from
destruction by random mutations. It is true that natural selection is
a preserving force that does a pretty good job at keeping beneficial
sequences "in memory" but only to a point. If you raise the mutation
rate too high, information is destroyed in a genome faster than the
reproduction rate and the forces of natural selection can keep up with
the loss and the entire population heads toward extinction. In fact,
in real life, this is exactly what is happening to the human gene
pool. Knightly and Crow have written a couple of papers estimating
that unless the average woman gives birth to over 40 offspring that
the human race will continue to head toward extinction.



> > Types of mutations:
> > Point, deletion, insertion, and recombination (note that all
> > deletions, insertions, and recombination mutations must be random with
> > respect to the number of characters involved as well as the position
> > of the mutation in the genome)
>
> This is exactly how the Word Mutagenator works.
>
> > Valid characters:
> > 26 English letters, period, and space (28 total)
> >
> > Beneficial selection:
> > Any string that has any portion of itself making meaningful sense in
> > the English language system will be accepted as "beneficial" in this
> > experiment. Those strings that have longer portions making meaningful
> > sense will be given a greater survival advantage ranging from 1 to
> > 1,000 advantage points.
>
> Any string or any portion?

Yes, as long as the portion that makes sense is sequential and it can
only be counted to be at a level of its total sequential length. For
example, if you have a string of 1,000 characters, but the longest
portion that actually makes meaningful sense is only 150 characters
long, you have only evolved to level 150. Make sense?

> This may not be necessary. If I bother, I'll
> probably require the entire string to make a complete thought in some sort
> of "pigeon" English.

It has to make sense according to Standard English usage in this
experiment - not some other English-like language or foreign language.



> > Starting point:
> > Each one of the 100 trillion genomes starts out with different
>
> If I endeavor to expand this project, I'll probably build something like the
> Mutagenator which has a population of exactly one for each genome, then
> mutates and recombines them randomly.

Again, a population is made up different genomes where each individual
already has a different genome from every other individual. Now, it
is true that different individuals may actually share the same genomic
sequence, but this is very unlikely with small populations at higher
levels of average sequence length.

Again, you cannot keep in sequences isolated from random mutation by
keeping them "in memory" as already discussed above. You must set up
your program to keep track of each individual sequence/genome in your
population as it evolves with the other sequences/genomes - just like
occurs in real life.



> > meaningful 7-character sequences (words or meaningful 7-character
> > phrases) repeated over and over again for a total of 1,000 characters.
>
> What's this all about? Why not start with "O"? Or let people pick their own
> words or phrases? Gotta make it a little interactive, you know.

That's fine. Pick whatever starting sequences you want as long as
they are less than 7-characters in meaningful length to start with and
as long as the total number of novel sequences/genomes in the
population does not go over the maximum predetermined colony and
genome size that your "computer environment" can sustain.



> > Note that you will be able to cover not only all meaningful
> > 7-character sequences in the English language system, but all possible
> > 7-character sequences in 7-character sequence space thousands of times
> > over.
> >
> > Goal:
> > Calculate how many generations it takes to evolve just 1,000
> > meaningful sequences at each level of complexity ranging from 1 to
> > 1,000 characters.
>
> I don't think you mean calculate, but generate.

I think is more like "keep track" of the number generations it took to
obtain a certain number of meaningful sequences at each level of
complexity once the meaningful sequences are "generated."

> If it is done randomly, then
> each run will be unique. Indeed, your requirement here makes little sense.

It makes a lot of sense because my argument is that it will take
exponentially more and more generations to evolve 1,000 meaningful
sequences at higher and higher levels of minimum sequence size. That
means that you have to keep track of the number of generations it took
to actually achieve the desired goal for each level of sequence
length.

> How 'bout we just generate phrases on the order of "to be or not to be", or
> "beware a war of words ere you err". That should prove the point
> sufficiently.

That is fine, but you must keep track of how many generations it took
you to generate or "evolve" phrases of such lengths compared to how
many generations it took to evolve equally meaningful phrases of
lesser and greater lengths.

> Or have we pushed your incredulity to such an extent that to push it further
> we have to directly model all 100 trillion genomes of 1000 bytes each for
> millions of generations?

You must model all of your genomes over the course of all generations,
whatever the size of your population. This is how other valid
computer generated models of evolution have done it (such as the one
performed by Lenski et. al.). You can use any size population you
want as long as the population does not exceed a predetermined size.



> > Obviously you will most likely start out covering
> > everything meaningful ranging from1 to 7 characters in size right at
> > your starting point. So, you can list all of these levels as taking
> > "zero generations" for your population to achieve success.
> >
> > It seems very likely to me
>
> Show your math.

What math would you like to see? The above estimate was based on a
starting population of genomes numbering 10^14. Since you can start
with whatever meaningful 7-letter sequence you want and since there
are most likely no more than 40,000 or so of these in the English
language system, 10^14 different sequences can easily cover 40,000
meaningful options billions of times over (about 2.5 billion, give or
take). Is that good enough for you? Or do you want more math?



> > that the next higher levels (i.e: 8, 9, 10,
> > etc) will take only one or two generations for your population to
> > evolve 1,000 uniquely meaningful sequences at each level.
>
> Now you want 1000 sequences of 1000 each? That's a megabyte. How about we
> make one, and then run the problem a thousand times.

That's fine, as long as you come up with a different sequence each
time. And, the way I have described the setup of the program, starting
the program over 1,000 times will be no different from just letting it
run since it will be like real evolution at every point in time.

And, if it makes that much difference to you, make the goal 10
sequences for each level so that we can get some sort of idea of the
_average_ generation time needed to achieve success at each level.
Just humor me here. If you can make it past the 100-character level
with even one genome at any point in time I will be impressed. But,
for the purposes of this experiment, official victory for you will be
achieved when you evolve 10 different 1,000-character sequences that
are meaningful in English according to English rules of structure and
grammar.

> > However, by
> > the time you get to level 25, I am thinking that your population is
> > going to start noticeably stalling in its ability to evolve the 1,000
> > uniquely meaningful English sequences. By level 50 I'm not sure that
> > your population of even 100 trillion will succeed in less than a
> > million generations.
>
> You keep going on about this 100 trillion. Well, there may be that many
> bacteria in your gut, but there isn't that much memory in your computer.

Certainly not, but even if there were, it wouldn't help you. So, good
luck with the much smaller population that your computer can actually
simulate. It is your claim that this really won't matter much, so
lets just see if you are correct.

> Please return to common-sense. We need to do this in thousands to make the
> computation feasible with the tools we have available.

Fine - it will only make your position all the more convincing if you
achieve success with a significantly smaller population.

> Keep in mind, this is your word-analogy. It was you who would not heed the
> warning to "beware a war of words." You said to start with short words and
> try to evolve longer words and that we would reach "walls" beyond which we
> couldn't go. This assertion has already been falsified, though you refuse to
> admit it.

LOL - how, exactly, has my assertion been falsified? There certainly
are walls beyond which evolution cannot go. Where have you falsified
this statement? It is certainly a falsifiable statement as any good
scientific hypothesis, but you certainly haven't falsified it. Not
even close. You haven't even demonstrated anything that I have not
already predicted. I never said that evolution would not work between
very short words and words that were only slightly less short. You
just implied that strawman assertion though you know full well that
wasn't even close to what I actually said. In fact, I actually said
that the walls beyond which evolution could not go were just a bit
farther away than those sequences just a dozen or two characters in
minimum size that you have claimed to "evolve" using real evolutionary
mechanisms. I also noted many times that the population size, genome
size, and mutation rate were very important to consider when
estimating the _average_ time needed for the evolution of new
functions at a given level of complexity.

What then have you proven counter to my position?



> > And, I feel fairly confident that by level 100
> > your population will do exceptionally well to succeed this side of a
> > trillion generations.
> >
> > Certainly you can see where I am going with this, but I'll do it
> > anyway. I predict that the generations needed to get to level 1,000
> > in this little game of evolution would be well over 10^1000 - truly
> > "zillions" of generations indeed!
> >
> > Now, see if you can prove that my predictions here are significantly
> > in error . . .
> >
> > Good luck with your "war of words." ; )
>
> Round one to Zachriel.

LOL - just keep saying that to yourself because anyone who actually
reads what my position is and has been for quite some time now will
know that you are full of nothing but hot air and bravado. You've
taken my statements out of context and built a strawman to simulate my
actual position. In fact, if anything you have only succeeded in
helping me in demonstrating the validity of my position, and for that
I do thank you. Most of those before you wouldn't even touch the idea
of comparing evolution in a human language system, like English or
computer code, to biological evolution. You, on the other hand, went
right out on that limb, and now you are about to find it cut out from
under your feet.

Round one - Zach sets himself up nicely to be knocked out in round
two.

> Round two may take a while.

Take your time. Make sure you do it right though and at least try
this time to attack my real position instead of some strawman
caricature of my position. You accuse me of moving my goalposts only
because you think to misrepresent where my goalposts have always been.

> I'll let you know if I have time to rewrite the
> program in a reasonable amount of time. The last project took a few weeks of
> spare time. We'll have to take it to the steering committee once the rules
> are hashed out.

I'm pretty busy myself or I would work on my own program for the
steering committee to review. Perhaps in late June or early July I
will have more time to do some programming of my own.

Good luck!

Sean
www.naturalselection.0catch.com

Zachriel

unread,
May 1, 2004, 10:51:54 PM5/1/04
to

"Sean Pitman" <seanpi...@naturalselection.0catch.com> wrote in message
news:80d0c26f.04050...@posting.google.com...

> "Zachriel" <sp...@zachriel.com> wrote in message
news:<voKdnSHbCpP...@adelphia.com>...
>
> > > Now, consider again your suggested population of 10^14 (100 trillion).
> > > With such a population you could cover all of the sequence space for
> > > 7-letter sequences 12,000 times over. Obviously then you are grossly
> > > misstating the strength of your argument.
> >
> > Um. In case you didn't know, I didn't directly simulate a population of
100
> > trillion. My genomes had a population size of just one, yes that's
> > right, just one for each species, the number of species limited to the
few
> > best (default 25), and their mutants which have at most one mutation
each.
>
> A population of "one for each species" is not a population of one, but
> many with "one for each species". Another thing, it seems like your
> program analyzed thousands of sequences before it chose the 25 that
> were "best".

1. Try a mutant.
2. If it is invalid discard it, and go to beginning.
3. If it is valid, then determine if it is "better" than any of the 25
existing strings.
4. If it is not "better", discard it, and go to beginning.
5. If it is "better", discard the weakest in the existing population, and
add the new word to the population, and go to beginning.

Validation checks mutants against the Dictionary; then if a valid word,
checks if they already exist in the population; then if they are new words,
inserts them in the population sorted by length. Only the longest are kept
to the next generation. (By the way, it doesn't generate then prune, but
prunes then generates. So when you stop the routine, you will see the
rejects at the end of the list as well as those that fit in the Pond. This
is so you can more easily see what is happening. You can press Prune to
eliminate them, or when you next Generate, you will see them being pruned
automatically.)


> If it were me, this is how I would set up the program: You start with
> a maximum steady state population limit as well as a genome size
> limit. In each generation you determine how many mutations, on
> average, each one of the members of your population can sustain. Now,
> the number of individuals in your population combined with the average
> mutation rate per individual gives you the maximum number of sequences
> that your population is allowed to analyze in a given generation. It
> didn't seem to me that you did this with your program. But, please do
> correct me if I am mistaken here.

Only one mutant at a time is considered, then resolved. This resembles how
life works. Signifiant mutations are rare, and as in life, most are
detrimental and immediately deselected. If you want to simulate the effects
of radiation and the radical breakdown of DNA, I'll leave that to you.


> > > The evolution of not just
> > > every meaningful word, but every _possible_ sequence in 7-letter
> > > sequence space could be achieved by your rather large population in
> > > very short order (i.e., less than one generation at population
> > > equilibrium within sequence space).
> >
> > It can be achieved by a very small population in a split second.
>
> Although this is a small point, it would be more meaningful to talk in
> generations here since modern computers can simulation huge numbers of
> generations, depending on population size, in a "split second". In
> any case, I would agree that 7-letter sequence evolution of quite a
> few meaningful 7-letter sequences can be achieved quite rapidly with a
> much smaller population than 100 trillion. I was just trying to point
> out to you what a population of 100 trillion could achieve at this
> relatively low level of meaningful sequence complexity.

It has to run on a standard desktop so that anyone can enjoy the program.


> > > I am also well aware that language systems, such as the English
> > > language system and even biological/genetic language systems tend to
> > > cluster meaningful symbols together. I am also aware that as you
> > > increase the minimum sequence length, these clusters start to separate
> > > from each other and become relatively smaller at the same time so that
> > > it becomes exponentially more and more difficult to evolve from one
> > > cluster to the next without having to cross through oceans of
> > > non-beneficial sequences.
> >
> > You have not shown that. Indeed, there is no way to know that from mere
> > mathematical analysis. You have to know their distribution. For all we
know,
> > words are all lined up nice and pretty in "permutation space". It turns
out
> > that many, perhaps most, of them are!
>
> Actually, you have not shown that either.

Sure it's been demonstrated. The Word Mutator evolves hundreds of new words
after considering a few thousand mutations. You had claimed that to evolve a
14-letter word would require looking for a few hundred words spread randomly
through a space of 109,418,989,131,512,359,209 possible permutations. Yet,
Word Mutator can find these words by searching with a process of mutation
and selection.
http://tinyurl.com/2p3z5


<snip>


>
> In real life, with the limit of a steady state population and genome
> size, once you evolve a new and "better" sequence, you loose the
> previous sequence. This is a problem, because it creates a rather
> constant limitation on the pool of options that each generation can
> use.

That is not necessarily true. In a living population, there are many copies
of each genome. If one should mutate then both can continue to exist. This
is exactly how organisms diversify. In Word Mutator, we keep the best of the
lot, which may include the original string, or not. The limiting factor is
the Pond size.


<snip>


>
> Again, a population is made up different genomes where each individual
> already has a different genome from every other individual. Now, it
> is true that different individuals may actually share the same genomic
> sequence, but this is very unlikely with small populations at higher
> levels of average sequence length.

Gee whiz Sean. Let's assume we start with a Pond that's full of clones of
"Sean". Then one of them mutates. This gives us a Pond that looks like this.

Sean
bean

So now we have a diverse gene-pool. If you want, you can input any number of
words into the Word Mutator and start with a diverse gene pool, or not. This
mutation and diversification is what the Word Mutator is all about.


> Again, you cannot keep in sequences isolated from random mutation by
> keeping them "in memory" as already discussed above. You must set up
> your program to keep track of each individual sequence/genome in your
> population as it evolves with the other sequences/genomes - just like
> occurs in real life.

If you have a billion copies of a particular organism, then some of those
organisms will have an unmutated genome, some will be mutated. So our
population starting with "sean" will after a while consist of "sean",
"bean", "sear", etc. "sean" will continue to exist unless it is deselected
by population pressures.


<snip>


>
> And, if it makes that much difference to you, make the goal 10
> sequences for each level so that we can get some sort of idea of the
> _average_ generation time needed to achieve success at each level.
> Just humor me here. If you can make it past the 100-character level
> with even one genome at any point in time I will be impressed.

It has nothing to do with "impressing" you. This is 93 characters:

"The origin of the life we know

Just like this poem rose from simple forms,

In meaning, and in kind, step-by-step."

This is 32 characters: "To be or not to be; that is the question."

Are you claiming these are not significantly meaningful strings of letters?


> But,
> for the purposes of this experiment, official victory for you will be
> achieved when you evolve 10 different 1,000-character sequences that
> are meaningful in English according to English rules of structure and
> grammar.

As I explained before, the number of possible mutants of thousand length
strings is a billion. Word Mutator can only do a few tens-of-thousand per
second, and Word Mutagenator is about ten times slower due to the Random
function. To run through enough mutations would take hours per generation.

Even at this slower speed, "pitman" evolved in about 5 minutes, after 144414
mutations (Pond = 50), into "reinstating" (11), "installing" (10), and a
whole lot of other words. Your calculaton would be like this:

E = 11-letter words in dictionary = 7100 ~ 10^4
P = possible 10-letter permutations = 26^11 = 10^15
average mutations required = E / P = one in a hundred billion.

And yet, it only took a hundred thousand. You're off by many orders of
magnitude. Please consider this discrepancy and tell us why your calcuations
are so far off.

If I do this project, it has to be able to run on a desktop computer so that
others can run the simulation themselves. Fortunately, it won't take as many
computations to prove the point. Nevertheless, you appear to be moving the
goal-posts so far that the project can't be complete in a reasonable period
of time.


<snip>


> > Keep in mind, this is your word-analogy. It was you who would not heed
the
> > warning to "beware a war of words." You said to start with short words
and
> > try to evolve longer words and that we would reach "walls" beyond which
we
> > couldn't go. This assertion has already been falsified, though you
refuse to
> > admit it.
>
> LOL - how, exactly, has my assertion been falsified? There certainly
> are walls beyond which evolution cannot go. Where have you falsified
> this statement? It is certainly a falsifiable statement as any good
> scientific hypothesis, but you certainly haven't falsified it. Not
> even close. You haven't even demonstrated anything that I have not
> already predicted. I never said that evolution would not work between
> very short words and words that were only slightly less short.

This was your assertion and your challenge:

"start a short 2 or 3-letter word and see how many words you can evolve that
require greater and greater minimum sequence requirements. No doubt you will
quickly find yourself coming to walls of meaningless or non-beneficial
potential options . . . "

You said to evolve "words". I did. The Word Mutator can evolve 10-letter
words starting from "O" in just seconds, and longer words in a few minutes.


> You
> just implied that strawman assertion though you know full well that
> wasn't even close to what I actually said. In fact, I actually said
> that the walls beyond which evolution could not go were just a bit
> farther away than those sequences just a dozen or two characters in
> minimum size that you have claimed to "evolve" using real evolutionary
> mechanisms. I also noted many times that the population size, genome
> size, and mutation rate were very important to consider when
> estimating the _average_ time needed for the evolution of new
> functions at a given level of complexity.
>
> What then have you proven counter to my position?

<snip>

This was your assertion and your challenge:

"start a short 2 or 3-letter word and see how many words you can evolve that
require greater and greater minimum sequence requirements. No doubt you will
quickly find yourself coming to walls of meaningless or non-beneficial
potential options . . . "

You said to evolve "words". I did. The Word Mutator can evolve 10-letter
words starting from "O" in just seconds, and longer words in a few minutes.


Zachriel

unread,
May 2, 2004, 12:01:18 AM5/2/04
to
<a clarification>

"Zachriel" <sp...@zachriel.com> wrote in message

news:uf6dnfggdOv...@adelphia.com...

> add the new word to the [next generation bin], and go to beginning.

The Word Mutator actually keeps all new words in a "next generation bin"
until after all the possible mutants from the last generation have been
considered. This process constitutes a "generation" and is according to the
posted Extended Rules. Only then is the pruning done and the next generation
started. This segregation between generations allows an accurate calculation
of the number of mutants available from a given starting point.

The Word Mutagenator inserts each mutant as it is created, but the actual
pruning is done in cycles for efficiency (at the time the screen is
updated).

Sean Pitman

unread,
May 2, 2004, 10:18:01 AM5/2/04
to
"Zachriel" <sp...@zachriel.com> wrote in message news:<uf6dnfggdOv...@adelphia.com>...

> > LOL - how, exactly, has my assertion been falsified? There certainly
> > are walls beyond which evolution cannot go. Where have you falsified
> > this statement? It is certainly a falsifiable statement as any good
> > scientific hypothesis, but you certainly haven't falsified it. Not
> > even close. You haven't even demonstrated anything that I have not
> > already predicted. I never said that evolution would not work between
> > very short words and words that were only slightly less short.
>
> This was your assertion and your challenge:
>
> "start a short 2 or 3-letter word and see how many words you can evolve that
> require greater and greater minimum sequence requirements. No doubt you will
> quickly find yourself coming to walls of meaningless or non-beneficial
> potential options . . . "
>
> You said to evolve "words". I did. The Word Mutator can evolve 10-letter
> words starting from "O" in just seconds, and longer words in a few minutes.

This statement in itself should give you a great deal of pause. You
yourself say that it takes only "seconds" to evolve short word
sequences up to 10-letters in size with your computer program, but
"minutes" to evolve longer sequences? Don't you see the exponential
expansion of required time in what you just said? This goes along
exactly with my predictions of what would take place with a
meaning-based selection process of mutating individuals. Very quickly
you will find that such an exponential expansion of required time to
completely stall out your computer's ability to evolve much of
anything beyond very very low levels of meaningful complexity
(relatively speaking).

From the context of all of these statements of mine that you quote,
you will easily see that the "walls" I am talking about that cannot be
crossed this side of eternity by any imaginable real life population
of organisms are over hundreds of "fairly specified amino acids in
size". In fact, many times I have drawn the line at a "couple
thousand fairly specified amino acids working together at the same
time". In the word analogy, I said that the same thing would be true.
You start with a short sequence just a few characters in size and,
with a good sized population and mutation rate, you can work your way
up to larger and larger sequences, but you will quickly find that the
evolutionary powers of your population start stalling out, until, at
very low levels of relatively complexity (i.e., less than 50 or so
highly specified characters depending upon population size and
mutation rate), your population simply cannot evolve anything
"beneficial" this side of a practical eternity of time.

Now you have said that my drawing the line at 50 or 100 is a dramatic
goal shift, but it is nothing of the sort. Relatively speaking, a
sequence with only 100 characters is nothing compared to the
meaningful sequence complexity that exists in say, a complete
Shakespearean play. The same is true of genetic sequences in
biological systems. A 1,000 amino acid protein system is nothing
compared to protein systems of function that require tens or even a
hundred thousand fairly highly specified amino acids working together
at the same time (i.e., not a cascade).

This is what I have said over and over again. You have just decided
to try and twist my position to some strawman version of what it
really is. The fact remains that if you start with a population of
short strings of say 2 or 3 characters in size and randomly mutate and
select them based on functional criteria that you will quickly find


yourself coming to walls of meaningless or non-beneficial potential

options. And, relatively speaking 50, 100, or even 1,000 character
sequences are extremely low level functions and so finding
insurmountable walls at such low levels is indeed a very "quick" stall
for evolutionary progression.

Again, where did I ever say otherwise outside of your false
insinuations?

Also, consider that I am being very generous with you here. I am
allowing the scenario you use to select for any meaningful word or
character sequence in the entire range of possible English language
possibilities without regard to their "beneficial" character for a
particular environment. In real life, not every meaningful genetic
sequence is beneficial. In fact, by far the greatest percent of
meaningful genetic sequences would be detrimental to a particular
organism in a particular environment.

Sean
www.naturalselection.0catch.com

Sean Pitman

unread,
May 2, 2004, 10:51:10 AM5/2/04
to
"Zachriel" <sp...@zachriel.com> wrote in message news:<EuydneL-jrU...@adelphia.com>...

> > > A population of "one for each species" is not a population of one, but
> > > many with "one for each species". Another thing, it seems like your
> > > program analyzed thousands of sequences before it chose the 25 that
> > > were "best".
> >
> > 1. Try a mutant.
> > 2. If it is invalid discard it, and go to beginning.
> > 3. If it is valid, then determine if it is "better" than any of the 25
> > existing strings.
> > 4. If it is not "better", discard it, and go to beginning.
> > 5. If it is "better", discard the weakest in the existing population, and
> > add the new word to the [next generation bin], and go to beginning.

This basically results in an extremely high reproductive rate as well
as mutation rate per organism in your population. If I recall
correctly, for one given type of string or individual "organism" in
your population, you allowed up to 95,000 mutant tries before you went
on to the next generation?! Is that correct? That translates into
95,000 offspring for one individual string with a mutation rate of 1
per offspring. That's truly an extraordinary reproduction rate and as
such does not correlate meaningfully with understanding the average
number of generations required to achieve success for more realistic
reproduction rates. Not even rapidly growing bacterial colonies have
such high individual reproduction rates, much less mutation rates. A
mutation rate of 1 per sequence length of only a few characters would
be lethal in real life, translating into millions of mutations per
average individual genome the size of a bacterial genome.

But, no matter. Even with such incredibly high reproduction rates and
mutation rates, your computer simulation will still find itself
rapidly stalling out in its ability to find new beneficial sequences
at higher and higher levels of meaningful English composition.

Oh, and by the way, by meaning I am talking about interactive meaning
where the sequence as a whole is greater in meaning than the sum of
its parts.

For example, the sequence, "Pizza haze storm grass" is not interactive
in standard English in that the words do not have a unified meaning
that is greater than their sum - as compared to the sequence, "Pizza
is my favorite food". You see how in the second sequence the meaning
of each individual portion of the sequence is linked to other portions
of the sequence to make a unified meaning that is much greater than
the sum of the smaller meanings of each smaller portion of the
sequence. If you remove part of the second sequence, you will loose
part or all of the unified meaning of the sequence. This is not so if
you remove part of the first sequence since the first sequence has no
readily apparent unified function.

> The Word Mutator actually keeps all new words in a "next generation bin"
> until after all the possible mutants from the last generation have been
> considered. This process constitutes a "generation" and is according to the
> posted Extended Rules. Only then is the pruning done and the next generation
> started. This segregation between generations allows an accurate calculation
> of the number of mutants available from a given starting point.

Again, you are looking at just the mutations available for one
individual here with a mutation rate of one and only one per
individual. If you look at the mutations available for the population
at large you will notice that there are far more possible mutations
per generation than you have loudly asserted on your website.

However, this really doesn't matter all that much other than it being
a point of clarification about what is really going on with your
program as compared to what happens in real-life organic evolution.

> The Word Mutagenator inserts each mutant as it is created, but the actual
> pruning is done in cycles for efficiency (at the time the screen is
> updated).

This part is just fine . . .

Sean
www.naturalselection.0catch.com

Sean Pitman

unread,
May 2, 2004, 11:01:05 AM5/2/04
to
"Zachriel" <sp...@zachriel.com> wrote in message news:<euqdnQ2wRbT...@adelphia.com>...

> > Beneficial selection:
> > Any string that has any portion of itself making meaningful sense in
> > the English language system will be accepted as "beneficial" in this
> > experiment.
>
> Um, how do you rigorously define "meaning"?

In order to be meaningful "as a whole" the entire sequence in question
must have interactive meaning where the sequence as a whole is greater


in meaning than the sum of its parts.

Consider that the character sequence, "pizza haze storm grass" is not


interactive in standard English in that the words do not have a
unified meaning that is greater than their sum - as compared to the
sequence, "Pizza is my favorite food". You see how in the second
sequence the meaning of each individual portion of the sequence is
linked to other portions of the sequence to make a unified meaning
that is much greater than the sum of the smaller meanings of each
smaller portion of the sequence. If you remove part of the second
sequence, you will loose part or all of the unified meaning of the
sequence. This is not so if you remove part of the first sequence
since the first sequence has no readily apparent unified function.

> Are these meaningful phrases?
>
> "dog"

Yes

> "a dog"

Yes

> "the cat barks"

Yes

> "I drown an eye"

Borderline, but I would still say yes.

> "soap, bread, milk, jalapeno peppers"

No

> "milk, flour, sugar, vanilla, baking powder"

No

> Does anyone have a meaning-o-meter I can borrow? It has to work fast,
> though, about a millisecond would be fine.

Perhaps a comparison to sequences used on the internet? - although I'm
not sure how fast this could be set up to work.

Sean
www.naturalselection.0catch.com

Zachriel

unread,
May 2, 2004, 11:58:37 AM5/2/04
to

"Sean Pitman" <seanpi...@naturalselection.0catch.com> wrote in message
news:80d0c26f.04050...@posting.google.com...

Except that your math is wrong. It takes longer to evolve complex genomes
than simple genomes. But you claim it is "zillions" and use fallacious
equations to back up your assertions.


> Very quickly
> you will find that such an exponential expansion

The number of mutants is on the order of L^3 where L is the length of the
string. You claim it is 26^L. These are very different numbers. Now consider
generations, G. You would claim that the number would be 26^L^G, but in
fact, with reasonable selection criteria, it is (L^3)*G. G is a product not
an exponent due to selection.

Your math is wrong, plain and simple


> of required time to
> completely stall out your computer's ability to evolve much of
> anything beyond very very low levels of meaningful complexity
> (relatively speaking).

Except that your math is wrong.


> From the context of all of these statements of mine that you quote,
> you will easily see that the "walls" I am talking about that cannot be
> crossed this side of eternity by any imaginable real life population
> of organisms are over hundreds of "fairly specified amino acids in
> size".

No, you referred specifically to walls between 2- and 3-letter words and
longer words. Let's quote you again: "start a short 2 or 3-letter word and
see how many WORDS you can evolve that require greater and greater minimum


sequence requirements. No doubt you will quickly find yourself coming to
walls of meaningless or non-beneficial potential options . . . "

If you wish to retract or modify this statement, that would be completely
acceptable. But after weeks of this, and a direct demonstration of the
fallacy of that statement, you have yet to retract it.


> In fact, many times I have drawn the line at a "couple
> thousand fairly specified amino acids working together at the same
> time".

You have also draw the line at long words. It does not take 26^L^G
permutations of discovery, but only on the order of (L^3)*G. In addition,
words are not distributed randomly, so our evolutionary exploration proceeds
quite rapidly. If words were distributed randomly, then our evolutionary
algorithm wouldn't work.


> In the word analogy, I said that the same thing would be true.
> You start with a short sequence just a few characters in size and,
> with a good sized population and mutation rate, you can work your way
> up to larger and larger sequences, but you will quickly find that the
> evolutionary powers of your population start stalling out, until, at
> very low levels of relatively complexity (i.e., less than 50 or so
> highly specified characters depending upon population size and
> mutation rate), your population simply cannot evolve anything
> "beneficial" this side of a practical eternity of time.

Let's quote you again. "start a short 2 or 3-letter word and see how many
WORDS you can evolve that require greater and greater minimum sequence


requirements. No doubt you will quickly find yourself coming to walls of
meaningless or non-beneficial potential options . . . "

How long of words did you have in mind? In addition, you are incorrect on
longer phrases. Every mutant of a 50 character string can be explored in
50^3 or 125000 mutations. That's it. It's not a Pitman Zillions. With a
reasonably defined selection algorithm, the total number of mutations
required is on the order of 125000*G.


> Now you have said that my drawing the line at 50 or 100 is a dramatic
> goal shift, but it is nothing of the sort. Relatively speaking, a
> sequence with only 100 characters is nothing compared to the
> meaningful sequence complexity that exists in say, a complete
> Shakespearean play.

Let's be specific again. You claimed that "the potential space of a
14-letter word or phrase is over 109,418,989,131,512,359,209 (over 100
million trillion)." This is very close to the Pitman Zillion and certainly
beyond the capability of any computer to explore in a reasonable time. My
Dictionary contains exactly 1643 (~10^3) 14-letter words. According to your
"calculations", my program would have to explore 10^19 / 10^3 or 10^16
mutants before having a decent chance of finding such a word. I invite
everyone to read your "analysis" of the problem for themselves.
http://tinyurl.com/2rx8g

And yet, the Word Mutator can solve this problem in minutes. How is it
possible? Can Sean Pitman please explain how the Word Mutator is capable of
doing this.

> The same is true of genetic sequences in
> biological systems. A 1,000 amino acid protein system is nothing
> compared to protein systems of function that require tens or even a
> hundred thousand fairly highly specified amino acids working together
> at the same time (i.e., not a cascade).

Garbage in, garbage out.


> This is what I have said over and over again. You have just decided
> to try and twist my position to some strawman version of what it
> really is. The fact remains that if you start with a population of
> short strings of say 2 or 3 characters in size and randomly mutate and
> select them based on functional criteria that you will quickly find
> yourself coming to walls of meaningless or non-beneficial potential
> options. And, relatively speaking 50, 100, or even 1,000 character
> sequences are extremely low level functions and so finding
> insurmountable walls at such low levels is indeed a very "quick" stall
> for evolutionary progression.

The answers are 10^5, 10^6 and 10^9 mutations per generation. If we can make
a reasonable selection for fitness at each generation, then we need just
multiply these numbers by the number of generations. In other words, we can
step through a hundred generations of a 50 character string in 10^7
mutations, and be able to find phrases of this form:

"Beware a war of words ere you err, Sean Pitman"


> Again, where did I ever say otherwise outside of your false
> insinuations?

"start a short 2 or 3-letter word and see how many WORDS you can evolve that


require greater and greater minimum sequence requirements. No doubt you will
quickly find yourself coming to walls of meaningless or non-beneficial
potential options . . . "

http://tinyurl.com/2gh7b

And "but the potential space of a 14-letter word or phrase is over
109,418,989,131,512,359,209 (over 100 million trillion)."
http://tinyurl.com/2rx8g

And where W = number of words of a given length and L is that length, then
inserting them into your calculation for 7-letters from the same post, "this
random walk will take, on average, over [ W/(26^L) ] mutations to arrive at
a new meaningful word at the level of [L]-letters.

So don't try and tell us that you didn't mean exactly that a 14-letter word
would be virtually impossible to evolve through the simple process of
mutation and selection for length. That's exactly what you meant. I have
noticed you have adjusted your rhetoric somewhat, but you have yet to
"clarify" your position. You need to do this as you have gone on about it
for months despite the protestations of many, many people on this newsgroup.

Zachriel

unread,
May 2, 2004, 12:23:08 PM5/2/04
to

"Sean Pitman" <seanpi...@naturalselection.0catch.com> wrote in message
news:80d0c26f.04050...@posting.google.com...

Now that's interesting. See below.


>
> > "I drown an eye"
>
> Borderline, but I would still say yes.

It's Shakespeare, but what does he know.


> > "soap, bread, milk, jalapeno peppers"
>
> No

It's obviously a shopping list.


>
> > "milk, flour, sugar, vanilla, baking powder"
>
> No

It's obviously a recipe.


> > Does anyone have a meaning-o-meter I can borrow? It has to work fast,
> > though, about a millisecond would be fine.
>
> Perhaps a comparison to sequences used on the internet? - although I'm
> not sure how fast this could be set up to work.


Phrases are composed of various structures. By identifying the allowable
structures, we might be able to come up with a reasonable algorithm.

noun + verb
adj + noun
adv + verb
verb + adv
art + noun
noun + conj
noun + conj + noun
verb + conj + verb
adj + noun + verb + adv
noun + conj + noun + verb
noun + verb + conj + noun + verb

" dog" is more meaningful than "dog".
"a dog" is more meaningful than " dog".
"a dog and" is more meaningful than "a dog".
"a dog and cat" is more meaningful still.

"a dog talks" is meaningful.
"a blue dog talks" is more meaningful.
"a blue dog talks jadedly"
"a blue dog and cat talk jadedly"
"a blue dog talks and a green cat drives"

And so on. Transitive verbs may be more difficult. It might be easier to
just let the computer randomly generate mutants, eliminating the obviously
invalid ones, then manually select the best.

Zachriel

unread,
May 2, 2004, 12:23:24 PM5/2/04
to

"Sean Pitman" <seanpi...@naturalselection.0catch.com> wrote in message
news:80d0c26f.04050...@posting.google.com...

Let's try explain this again. Each word represents a species. Each species
is subject to occasional mutation. We consider each of these possible
mutations and determine if they make a valid word; if they do, then we
determine if it is longer than others in our population. We continue this
process and our population will change over time, getting longer and longer.

In Word Mutator, every single mutation and recombination is considered. This
is according to the Extended Rules as posted earlier in the thread. In Word
Mutagenator, random mutation of random words and recombination of random
snippets at random points in random words is used. Every possible mutation
and recombination has an equally likely chance of being considered.

And in both cases, our population of words will change over time, getting
longer and longer.


> But, no matter. Even with such incredibly high reproduction rates and
> mutation rates, your computer simulation will still find itself
> rapidly stalling out in its ability to find new beneficial sequences
> at higher and higher levels of meaningful English composition.

The Word Mutator doesn't handle phrases, only words. It was meant as a
response to this specific claim and this challenge.

"start a short 2 or 3-letter word and see how many WORDS you can evolve that


require greater and greater minimum sequence requirements. No doubt you will
quickly find yourself coming to walls of meaningless or non-beneficial
potential options . . . "

http://tinyurl.com/2gh7b

Well, I did. And it turns out I can evolve all sorts of words, some long,
some short, some with high Scrabble score, some combination of these, an
assigned preference quotient, or just about any rule we choose to use.


>
> Oh, and by the way, by meaning I am talking about interactive meaning
> where the sequence as a whole is greater in meaning than the sum of
> its parts.

And exactly how do you suppose we can program a computer to select for that?
So you have indeed moved the goal-posts at supersonic speeds.

Another specific claim. Time to put up, Sean. Please put "pitman" or
whatever you want into the Word Mutator and find a mutant--according to the
posted rules--that is not accounted for. Don't make claims for which you
have no support. The code is open source. The possible mutations are clearly
defined.

For a starting population of "sean" and "pitman" there are 31 possible valid
words out of at most 1014 mutants. The words are (in order by scrabble
score), "jean" with a scrabble score of 11, plus "pitman, seaman, swan,
wean, yean, bean, mean, scan, seam, span, sedan, man, pan, pin, pit, dean,
ma, pi, lean, seal, sean, sear, seas, seat, seen, sea, an, it, a, i" The
second generation, we find "zeal" at 13, the third generation, "mazy" at 18,
plus "sean" was rediscovered after the original had been eliminated with a
special bonus Scrabble score of 99 (as set in the Dictionary). Then
"zipping" at 21, then "pickaxes" at 23, and so on, but "sean" still takes
the cake at 99.

In any case, the Word Mutator is capable of finding 14-letter words in a
"sequence space of hundreds of trillions and does it by only looking at a
few thousand mutations. Please explain how this is possible.

Sean Pitman

unread,
May 3, 2004, 8:49:47 PM5/3/04
to
"Zachriel" <sp...@zachriel.com> wrote in message news:<Duqdnd5d3_P...@adelphia.com>...

> > > Are these meaningful phrases?
> > >
> > > "dog"
> >
> > Yes
> >
> > > "a dog"
> >
> > Yes
> >
> > > "the cat barks"
> >
> > Yes
>
> Now that's interesting. See below.
>
> > > "I drown an eye"
> >
> > Borderline, but I would still say yes.
>
> It's Shakespeare, but what does he know.

Shakespeare! I repent in dust and ashes! ; )



> > > "soap, bread, milk, jalapeno peppers"
> >
> > No
>
> It's obviously a shopping list.

This sequence is not obviously composed of interdependent subparts and
therefore it is not more complex than its most complex subpart. In my
book, it receives a maximum selectability score of 16 or the
meaningful combination of "jalapeno peppers".

> > > "milk, flour, sugar, vanilla, baking powder"
> >
> > No
>
> It's obviously a recipe.

Again, this sequence is not _obviously_ composed of interdependent
subparts so it looses selectability points and has a maximum
selectability score of 13 for "baking powder".



> Phrases are composed of various structures. By identifying the allowable
> structures, we might be able to come up with a reasonable algorithm.
>
> noun + verb
> adj + noun
> adv + verb
> verb + adv
> art + noun
> noun + conj
> noun + conj + noun
> verb + conj + verb
> adj + noun + verb + adv
> noun + conj + noun + verb
> noun + verb + conj + noun + verb

Most of the time I would say that just about anything I can think of
that follows these rules would work as "meaningful" in the English
language system. Not necessarily "beneficial" in a given environment,
but probably meaningful. And, since we have decided not to worry
about the idea of "beneficial" for the purposes of this experiment, I
think that this setup will probably do. But certainly, I will not
accept your "laundry list" of words as having a selectability level
equal to its total length.

> " dog" is more meaningful than "dog".

Ok, I'll go along with that.

> "a dog" is more meaningful than " dog".
> "a dog and" is more meaningful than "a dog".
> "a dog and cat" is more meaningful still.

I'll agree with that. Maximum selectability level = 13 for the "a dog
and cat" phrase.

> "a dog talks" is meaningful.
> "a blue dog talks" is more meaningful.
> "a blue dog talks jadedly"
> "a blue dog and cat talk jadedly"
> "a blue dog talks and a green cat drives"

Yep. Maximum selectability level = 39 for the "a blue dog talks and a
green cat drives" sentence.

> And so on. Transitive verbs may be more difficult. It might be easier to
> just let the computer randomly generate mutants, eliminating the obviously
> invalid ones, then manually select the best.

Actually no. Manual selection is not allowed at all in this scenario
under any circumstances since it entails intelligent design via the
use of insight and pre-determined futuristic goals. A _mindless_
Nature cannot select based on some future potential function. Nature
can only select based on what works right here and now. Of course, I
know you really like to do the "manual selection" thing since that
dramatically narrows the field of options and significantly speeds up
the evolutionary process. But, this simply is not allowed since a
mindless nature does not have access to this ability and that is
exactly what we are trying to model - the abilities of a _mindless_
non-intelligent natural process.

Also, sequence eliminations will not necessarily eliminate the
"obviously invalid" sequences. The computer must be programmed to
only eliminate sequences from the offspring based on those that are
not in the top number of predetermined sequences with the highest
selectability scores (as detailed below). The actual number of
positively selected sequences must not have an average number greater
than the pre-determined steady-state population number (also detailed
below).

My suggested parameters for this experiment are as follows:

Population Size:
Anything maximum population size that you think your computer can
easily handle - perhaps a steady state of 100
individuals/genomes/sequences?

Types of Sequence Characters:
All 26-letters of the English alphabet plus whatever punctuation marks
you wish to include, such as a space, period, comma, etc . . .

Reproduction Rate:
I think it would be good if we at least tried to make this one at
least somewhat realistic. For example, lets limit the average
reproduction rate per individual in our population to less than 100
offspring per generation. Fair enough? For example, this means that
a population of 100 different genomes will not collectively produce
more than 10,000 different genomes/offspring in each generation on
average.

Selection:
Any sequence or any portion of a sequence (made up of single or
multiple words) that is meaningful according to standard rules of
English usage and grammar may be selected as advantageous relative to
its peers. Those sequences having a higher sequence score will be
rewarded accordingly by being allowed to produce an equivalently
greater number of offspring in the next generation, though the total
number of offspring will not exceed the above stated limit on average
over the course of the generations.

Sequence Value:
The sequence with the longest consecutive internal sequence that has
unified meaning will receive the highest scores. For example,
consider the sequence, "toy, tree, ear, glove, run". This sequence
does not have a unified meaning that is greater than the sum of all of
its internal parts. Since the longest part of this sequence that does
carry a complete unified meaning is only 5-characters in length, the
maximum selectability score of this sequence is only "5". On the
other hand, consider the sequence, "See the little boy play in the
dirt." This sequence does have a unified meaning that is provided by
all parts of the sequence working together. Therefore, this sequence
gets a selectability score of "36", which is far greater than the
score of "5" earned by the first sequence. However, if there were no
sequences with a score higher than "5" in the population, the first
sequence would still be the most selectable. But, just because it has
5 obviously meaningful words does not make it more selectable than an
apparently more random string that also has a complexity score of "5",
such as "wh okras irhoijtc tizmp". Although it seems like most of
this sequence has no meaning at all, there are several portions of it
that do have English-language meaning. Internal sequences like "ok"
and "as" and "tiz" and "ho" all have collective individual
English-language meaning. However, the longest internal meaningful
sequence is "okras". Since this is the longest internal sequence with
meaning, the sequence as a whole would also get a complexity score of
"5". Of course, as previously mentioned, the higher the score, the
more offspring will be produced by that sequence relative to its 100
peers in the population.

Types of Mutations:
Each type of the following listed mutations can be given whatever
weight value of occurrence you want - although in real life point
mutations are far more common than certain other types of mutations,
such as recombination mutations. But, that doesn't really make much
of a difference for the purpose of this experiment.

Point mutations - A single character change in one position of a
character sequence.

Deletion mutations - The loss of one or more character positions
from an individual genome. The number of characters lost must be
random per deletion mutation..

Insertion mutations - The random insertion of one or more random
characters at a random position within an individual genome. The
number of characters inserted must also be random per insertion
mutation.

Recombination mutations - Any sequence may randomly recombine at a
random location within its own genome with any other sequence,
randomly chosen, at a random site within that genome. The
recombination must be balanced between the two recombining sequences
and may destroy the meaning of a previously meaningful word or phrase
in the genomes of one or both of the involved genomes.

Cut and paste mutations - A random portion of any genome (not
limited to intact words or meaningful sequences), chosen at random, my
be cut out and pasted into another genome, chosen at random, in a
random location (not limited to certain ideal locations at the
beginning or ends of intact words or phrases). In other words, the
cut and paste mutation could destroy the meaning of a previously
meaningful word or phrase in the genomes of one or both of the
involved genomes.

NOTE - pay special attention to the definitions of recombination and
cut and paste mutations listed here since you did not program you
computer to work like this. Instead, you programmed your computer to
always select fully intact meaningful sequences (or "words" in your
case) to insert into other words at various places. Although this can
happen in real life, real life mutations do not have to work like this
and in fact usually do not work like this. Allowing for partial
recombination and copying of sequence of origin greatly increases the
average time required to achieve a meaningful mutation event.

Mutation Rate:
As long as the other rules listed here are followed, the average
mutation rate per individual genome length can be pretty much anything
you want it to be. Remember though that if the mutation rate gets too
high, it will end up destroying higher levels of meaningful complexity
faster than they can be built and your population as a whole will head
downhill in complexity. Also note that the mutation rate is an
_average_ rate per given length of a genome. This means that it is
indeed possible for a higher number of mutations to affect a given
region of a genome in a given offspring in a given generation.

Starting Point:
Start with the above-determined number of individual genomes (lets say
100 for now, but it can be whatever you want) in your steady state
population. They can all be the same exact sequence or they can each
be very different sequences. Just for kicks, lets say that they must
all be worth less than 3 selectability points to start out with. They
can be short or long to start with, it really doesn't matter. A given
sequence may even be made up of a long stretch of just one letter
repeated over and over again. For example, the sequence,
"AAAAAAAAAAAAA" would be accepted as valid and would have a
selectability score of 1 point since the longest meaningful internal
sequence in this hypothetical genome is the English word, "A".

Generations Required to Reach each Level of Complexity:
Keep track of the number of generations it takes your population as a
whole to reach each level of complexity as defined above.

Winning the Game:
You will win the game if and when your population evolves a meaningful
English language sequence in any one of its individual genomes that is
worth just 1,000 selectability points as defined above. Coming short
of this level, I will reward you with major brownie points for
achieving the level of just 100 selectability points and even raise an
eyebrow if you reach 50 selectability points.

Good luck to you! May all the forces of evolution and mindless
creativity be with you!

Sean
www.naturalselection.0catch.com

Zachriel

unread,
May 3, 2004, 10:05:06 PM5/3/04
to

"Sean Pitman" <seanpi...@naturalselection.0catch.com> wrote in message
news:80d0c26f.04050...@posting.google.com...
> "Zachriel" <sp...@zachriel.com> wrote in message
news:<Duqdnd5d3_P...@adelphia.com>...
>
> > > > Are these meaningful phrases?
> > > >
> > > > "dog"
> > >
> > > Yes
> > >
> > > > "a dog"
> > >
> > > Yes
> > >
> > > > "the cat barks"
> > >
> > > Yes
> >
> > Now that's interesting. See below.
> >
> > > > "I drown an eye"
> > >
> > > Borderline, but I would still say yes.
> >
> > It's Shakespeare, but what does he know.
>
> Shakespeare! I repent in dust and ashes! ; )
>
> > > > "soap, bread, milk, jalapeno peppers"
> > >
> > > No
> >
> > It's obviously a shopping list.
>
> This sequence is not obviously composed of interdependent subparts and
> therefore it is not more complex than its most complex subpart. In my
> book, it receives a maximum selectability score of 16 or the
> meaningful combination of "jalapeno peppers".

They are related. They are all things you buy at the grocery store.

> > > > "milk, flour, sugar, vanilla, baking powder"
> > >
> > > No
> >
> > It's obviously a recipe.
>
> Again, this sequence is not _obviously_ composed of interdependent
> subparts so it looses selectability points and has a maximum
> selectability score of 13 for "baking powder".

They certainly do relate. They are all ingredients of a cake. They are all
food. They are more related than "milk, cars, atoms, Toledo". However, how
to quantize this, I don't really know. I think you will find that lists have
an important function, too, but it is not essential to the overall project.


> > Phrases are composed of various structures. By identifying the allowable
> > structures, we might be able to come up with a reasonable algorithm.
> >
> > noun + verb
> > adj + noun
> > adv + verb
> > verb + adv
> > art + noun
> > noun + conj
> > noun + conj + noun
> > verb + conj + verb
> > adj + noun + verb + adv
> > noun + conj + noun + verb
> > noun + verb + conj + noun + verb
>
> Most of the time I would say that just about anything I can think of
> that follows these rules would work as "meaningful" in the English
> language system. Not necessarily "beneficial" in a given environment,
> but probably meaningful. And, since we have decided not to worry
> about the idea of "beneficial" for the purposes of this experiment, I
> think that this setup will probably do. But certainly, I will not
> accept your "laundry list" of words as having a selectability level
> equal to its total length.

Thank you for accepting these simple grammar rules.


> > " dog" is more meaningful than "dog".
>
> Ok, I'll go along with that.
>
> > "a dog" is more meaningful than " dog".
> > "a dog and" is more meaningful than "a dog".
> > "a dog and cat" is more meaningful still.
>
> I'll agree with that. Maximum selectability level = 13 for the "a dog
> and cat" phrase.

I still don't understand your scoring system.


> > "a dog talks" is meaningful.
> > "a blue dog talks" is more meaningful.
> > "a blue dog talks jadedly"
> > "a blue dog and cat talk jadedly"
> > "a blue dog talks and a green cat drives"
>
> Yep. Maximum selectability level = 39 for the "a blue dog talks and a
> green cat drives" sentence.
>
> > And so on. Transitive verbs may be more difficult. It might be easier
to
> > just let the computer randomly generate mutants, eliminating the
obviously
> > invalid ones, then manually select the best.
>
> Actually no. Manual selection is not allowed at all in this scenario
> under any circumstances since it entails intelligent design via the
> use of insight and pre-determined futuristic goals.

As you have accepted the proposed grammar rules, we don't need a human
meaning-o-meter. These rules are somewhat simplified, of course, and I will
expand on them.


> A _mindless_
> Nature cannot select based on some future potential function. Nature
> can only select based on what works right here and now. Of course, I
> know you really like to do the "manual selection" thing since that
> dramatically narrows the field of options and significantly speeds up
> the evolutionary process.

Actually not. I want the computer to do it. There are two technical issues I
have. I need a text word-list with the grammar parts of each word. The
longest I have found is only 2000 word long.
http://www1.harenet.ne.jp/~waring/vocab/wordlists/1-2000.txt

The other issue is building the grammar parser. There are issues with tense
and plurality. I might just make everything present tense singular.

Sean Pitman

unread,
May 5, 2004, 10:26:09 AM5/5/04
to
"Zachriel" <sp...@zachriel.com> wrote in message news:<Ndqdnd85csL...@adelphia.com>...

> "Sean Pitman" <seanpi...@naturalselection.0catch.com> wrote in message

> > > > > "soap, bread, milk, jalapeno peppers"


> > > >
> > > > No
> > >
> > > It's obviously a shopping list.
> >
> > This sequence is not obviously composed of interdependent subparts and
> > therefore it is not more complex than its most complex subpart. In my
> > book, it receives a maximum selectability score of 16 or the
> > meaningful combination of "jalapeno peppers".
>
> They are related. They are all things you buy at the grocery store.

For one thing, that is not _necessarily_ so and this limitation is not
clear from the statement itself. Also, this sequence is not very
highly "specified". What I mean by this is that the order of the
words in this phrases does not obviously lend itself to a greater
collective meaning of the overall sequence. For example, I could
switch the order that "bread" and "milk" occur in the list so that it
reads, "soap, milk, bread, jalapeno peppers", and the overall meaning
of the sequence would not obviously change. I could do this with any
word in this sequence. This is not allowed in this scenario. The
overall sequence must have a greater collective meaning than the sum
of its parts. Your grocery store list does not follow this rule. It
has the same, or at least not that much greater, collective meaning as


the sum of its parts.

> > > > > "milk, flour, sugar, vanilla, baking powder"


> > > >
> > > > No
> > >
> > > It's obviously a recipe.
> >
> > Again, this sequence is not _obviously_ composed of interdependent
> > subparts so it looses selectability points and has a maximum
> > selectability score of 13 for "baking powder".
>
> They certainly do relate. They are all ingredients of a cake.

They do not form an interdependent internal structure. Therefore,
this sequence as a whole is not significantly greater in complexity
than the sum of its parts. It's maximum selectablity score is
therefore based only on the longest internal sequence that does show
an interdependent internal structure. In this case, the longest
internal sequence is "baking powder", which is 13 characters long and
so receives 13 selectabilty points.



> > > Phrases are composed of various structures. By identifying the allowable
> > > structures, we might be able to come up with a reasonable algorithm.
> > >
> > > noun + verb
> > > adj + noun
> > > adv + verb
> > > verb + adv
> > > art + noun
> > > noun + conj
> > > noun + conj + noun
> > > verb + conj + verb
> > > adj + noun + verb + adv
> > > noun + conj + noun + verb
> > > noun + verb + conj + noun + verb
> >
> > Most of the time I would say that just about anything I can think of
> > that follows these rules would work as "meaningful" in the English
> > language system. Not necessarily "beneficial" in a given environment,
> > but probably meaningful. And, since we have decided not to worry
> > about the idea of "beneficial" for the purposes of this experiment, I
> > think that this setup will probably do. But certainly, I will not
> > accept your "laundry list" of words as having a selectability level
> > equal to its total length.
>
> Thank you for accepting these simple grammar rules.

You're welcome, and good luck.



> > > " dog" is more meaningful than "dog".
> >
> > Ok, I'll go along with that.
> >
> > > "a dog" is more meaningful than " dog".
> > > "a dog and" is more meaningful than "a dog".
> > > "a dog and cat" is more meaningful still.
> >
> > I'll agree with that. Maximum selectability level = 13 for the "a dog
> > and cat" phrase.
>
> I still don't understand your scoring system.

My selectability score is based on the longest internal sequence that
has an interdependent internal structure that gives a higher
collective meaning in the English language system. For example, all
the words in a sentence support each other in a common goal or theme.
All the sentences in a meaningful paragraph also relate to every other
sentence in that paragraph, creating their own "beneficial"
environment. If a sentence does not have anything obviously to do
with the paragraph in question, then it is not allowed to be part of
that paragraph.



> > > "a dog talks" is meaningful.
> > > "a blue dog talks" is more meaningful.
> > > "a blue dog talks jadedly"
> > > "a blue dog and cat talk jadedly"
> > > "a blue dog talks and a green cat drives"
> >
> > Yep. Maximum selectability level = 39 for the "a blue dog talks and a
> > green cat drives" sentence.
> >
> > > And so on. Transitive verbs may be more difficult. It might be easier
> > > to just let the computer randomly generate mutants, eliminating the
> > > obviously invalid ones, then manually select the best.
> >
> > Actually no. Manual selection is not allowed at all in this scenario
> > under any circumstances since it entails intelligent design via the
> > use of insight and pre-determined futuristic goals.
>
> As you have accepted the proposed grammar rules, we don't need a human
> meaning-o-meter. These rules are somewhat simplified, of course, and I will
> expand on them.

You could actually use human selection according to certain rules that
did not involve futuristic knowledge or goals, but only involved the
selection of current meaning or function for a sequence as compared to
other sequences. In other words, the human would have to follow the
same rules that the computer is programmed to follow (as previously
detailed), but which the computer may be too slow to follow. However,
it seems that this would probably be much less practical than seeing
if you could get the computer to do everything to start with.



> > A _mindless_
> > Nature cannot select based on some future potential function. Nature
> > can only select based on what works right here and now. Of course, I
> > know you really like to do the "manual selection" thing since that
> > dramatically narrows the field of options and significantly speeds up
> > the evolutionary process.
>
> Actually not. I want the computer to do it. There are two technical issues I
> have. I need a text word-list with the grammar parts of each word. The
> longest I have found is only 2000 word long.
> http://www1.harenet.ne.jp/~waring/vocab/wordlists/1-2000.txt
>
> The other issue is building the grammar parser. There are issues with tense
> and plurality. I might just make everything present tense singular.

That may work, but it would create another limit that would narrow the
field that much more rapidly as your computer tries to evolve up the
ladder of complexity.

Sean

Sean Pitman

unread,
May 5, 2004, 2:09:58 PM5/5/04
to
"Zachriel" <sp...@zachriel.com> wrote in message news:<YNidnTkRdL0...@adelphia.com>...

> > This statement in itself should give you a great deal of pause. You
> > yourself say that it takes only "seconds" to evolve short word
> > sequences up to 10-letters in size with your computer program, but
> > "minutes" to evolve longer sequences? Don't you see the exponential
> > expansion of required time in what you just said? This goes along
> > exactly with my predictions of what would take place with a
> > meaning-based selection process of mutating individuals.
>
> Except that your math is wrong. It takes longer to evolve complex genomes
> than simple genomes. But you claim it is "zillions" and use fallacious
> equations to back up your assertions.

Actually I think that it is your understanding of what is really going
on, not so much your math, that is wrong. Your math is pretty much
correct, but you are not working on the right problem. It is in
thinking that you are working on the right problem that your math is
incorrect.



> > Very quickly
> > you will find that such an exponential expansion
>
> The number of mutants is on the order of L^3 where L is the length of the
> string.

Here is where you make one of your mistakes in conceptualizing the
problem. The total number of possible mutants is C^L where C is the
number of available characters and L is the length of the string.
That IS the total number of possible mutants for a character string of
a certain length. There just is no way around this concept.

You, of course, are trying to see how many possible mutants exist
within just one mutational step away from one particular string. That
is basically a meaningless number with regard to the problem at hand.
In fact, in real life a given generation cannot analyze and subject to
a selection process all 1-step possibilities as you are doing with
your computer program. In real life the number of new sequences that
can be analyzed is limited by the reproductive rate of that sequence.
If the sequence only produces 5 offspring, only 5 new sequences can be
analyzed, regardless of the fact that say 20,000 novel sequences could
have been produced in that generation.

It doesn't matter how many total unique sequences are within one step
of one string. What matters is the average density of "beneficial"
sequences or the likelihood that one of these beneficial sequences is
just one step away from the current sequence. The only way you can
raise your odds of finding a "beneficial" sequence in each generation,
is by increasing the reproduction rate and/or the steady state
population size.

For example, if your steady state population is just 2 and your
reproductive rate is also 2 per individual, your population as a whole
can only search a maximum of 4 new sequences in sequence space in each
generation (with an average mutation rate that is greater than 1 per
genome). After each of the 2 individuals give rise to a total of 4
offspring, these 4 offspring are subjected to the forces of selection,
which will eliminate 2 of them, on average, to keep the steady state
population at 2. If there is no selectable difference between these 4
individuals, the elimination process will be purely random. However,
say that one of these 4 sequences happens to hit a sequence that
caries a higher selectability score than the sequences of its peers.
In such a situation, this sequence would be preferentially selected
over its 3 other peers to populate the next generation.

So you see, with such a low reproductive rate and steady state
population size, it would take a lot longer to find a beneficial
sequence that averaged, say 1 in 100 sequences at a particular level
of complexity compared to a larger population with a higher
reproductive rate. For example, if your population where made up of
10 novel individual sequences and your reproductive rate was also 10
offspring per generation, the maximum number of new sequences that
could be evaluated per generation would be a lot higher, at 100, as
compared to the first scenario, at only 4.

Make sense?

> You claim it is 26^L. These are very different numbers. Now consider
> generations, G. You would claim that the number would be 26^L^G

This is not true at all. If you consider again what I wrote above, I
would claim that the number of searchable sequences per generation is
the number of offspring produced per generation when the mutation rate
is greater than one. Of course, this is 26^L^G. Far from it.

> but in
> fact, with reasonable selection criteria, it is (L^3)*G. G is a product not
> an exponent due to selection.

This is a meaningless formula, as described above. The maximum
searchable sequences over the course of a certain number of
generations is determined by the average number of offspring per
generation times the number of generations. It has nothing at all to
do with sequence length.

> Your math is wrong, plain and simple

My math is not wrong and yours is completely irrelevant to the actual
problem.



> > of required time to
> > completely stall out your computer's ability to evolve much of
> > anything beyond very very low levels of meaningful complexity
> > (relatively speaking).
>
> Except that your math is wrong.

See above . . .



> > From the context of all of these statements of mine that you quote,
> > you will easily see that the "walls" I am talking about that cannot be
> > crossed this side of eternity by any imaginable real life population
> > of organisms are over hundreds of "fairly specified amino acids in
> > size".
>
> No, you referred specifically to walls between 2- and 3-letter words and
> longer words. Let's quote you again: "start a short 2 or 3-letter word and
> see how many WORDS you can evolve that require greater and greater minimum
> sequence requirements. No doubt you will quickly find yourself coming to
> walls of meaningless or non-beneficial potential options . . . "
>
> If you wish to retract or modify this statement, that would be completely
> acceptable. But after weeks of this, and a direct demonstration of the
> fallacy of that statement, you have yet to retract it.

That is because the statement, as it stands, is not wrong at all. I
would say the very same thing again. At each level of complexity,
starting with 1, 2, or 3-letter words or sequences, the walls between
the current level and the next higher level begin to grow in size - in
an exponential fashion, until, relatively quickly, they become
insurmountable for a given steady state population with a given
reproductive rate and mutation rate this side of a practical eternity
of time. This is what I've always said and you have not countered
this statement at all. It is just as valid as it ever was. The only
thing you thought you could get away with is claiming that I said that
the growing walls became insurmountable at very low levels of
meaningful sequence complexity even in the face of large populations
with extraordinarily high reproductive rates and mutation rates. Now,
that is just a silly argument and it really should be removed from
your website. It is nothing but a strawman version of reality and you
know it.



> > In the word analogy, I said that the same thing would be true.
> > You start with a short sequence just a few characters in size and,
> > with a good sized population and mutation rate, you can work your way
> > up to larger and larger sequences, but you will quickly find that the
> > evolutionary powers of your population start stalling out, until, at
> > very low levels of relatively complexity (i.e., less than 50 or so
> > highly specified characters depending upon population size and
> > mutation rate), your population simply cannot evolve anything
> > "beneficial" this side of a practical eternity of time.
>
> Let's quote you again. "start a short 2 or 3-letter word and see how many
> WORDS you can evolve that require greater and greater minimum sequence
> requirements. No doubt you will quickly find yourself coming to walls of
> meaningless or non-beneficial potential options . . . "
>
> How long of words did you have in mind? In addition, you are incorrect on
> longer phrases. Every mutant of a 50 character string can be explored in
> 50^3 or 125000 mutations. That's it. It's not a Pitman Zillions.

Every mutant of a 50-character string cannot be explored in 50^3
mutations. That is a demonstrateably ludicrous statement and I am
surprised that you are still using it. The possible permutations of a
50-character string are 26^50 or around 10^70. Again, this has
already be discussed above.

> With a
> reasonably defined selection algorithm, the total number of mutations
> required is on the order of 125000*G.

This number says nothing about the average number of mutations
necessary to find the first _meaningful_ sequence in the next higher
level of complexity. It says nothing at all about this. This is only
the sequence space that is within one step of a 50-character string,
which has nothing at all to do with the odds that one of these
surrounding sequences will be beneficially selectable. I honestly
don't know how you think your calculations are remotely relevant.

Sean
www.naturalselection.0catch.com

Zachriel

unread,
May 5, 2004, 10:17:46 PM5/5/04
to

"Sean Pitman" <seanpi...@naturalselection.0catch.com> wrote in message
news:80d0c26f.04050...@posting.google.com...

That is incorrect. The rules for mutation were clearly defined. Do you now
want to add additional or different rules of mutation? We have
point-mutations, delete mutations, insert mutations, snips, remainers and
recombinations.

"Just try a little experiment yourself. Start with a short 2 or 3-letter
word and see how many words you can evolve . . . "
http://tinyurl.com/33otq

Please tell us, how many valid mutations can be made to this word, "at"? How
many of these make valid words?


> You, of course, are trying to see how many possible mutants exist
> within just one mutational step away from one particular string. That
> is basically a meaningless number with regard to the problem at hand.

<snip>

That's the Sean Pitman Word Analogy, not mine.

Zachriel

unread,
May 6, 2004, 7:21:09 AM5/6/04
to

"Zachriel" <sp...@zachriel.com> wrote in message
news:_sydnWc3S7X...@adelphia.com...

<from an earlier post>


> > There are two technical issues I have.
> > I need a text word-list with the grammar parts of each word.
> > The longest I have found is only 2000 word long.
http://www1.harenet.ne.jp/~waring/vocab/wordlists/1-2000.txt

Words of wisdom about water sports from Sean Pitman:
"If I want to evolve a new 7-letter word starting with meaningful 7-letter
word, I will have to swim through this ocean of meaningless words."
http://tinyurl.com/2ju8q


So I thought to myself, what if there were only 2,000 words in our
Dictionary instead of 80,000. That would make the Pitman Gaps between words
even larger. Certainly then, Mutagenation would fail.

I plugged in the 2,000 most common words in the English language. There were
actually only 1841 words without duplicates. And guess what! Mutgenation had
absolutely no problem "swimming through the ocean of meaningless words"!

jpg of word distribution
http://www.zachriel.com/mutagenation/2000words.jpg

jpg of Word Mutagenator after about 3 minutes (Pond = 100)
http://www.zachriel.com/mutagenation/1841mutagens.jpg
114 word of length 9
Pitman Number(9) = 47,627,225,254

jpg of Word Mutator after about 3 minutes (Pond = 100)
http://www.zachriel.com/mutagenation/1841mutants.jpg
48 words of length 11
Pitman Number(11) = 76,465,510,145,579

And it didn't take a zillion years (or even a zillion seconds)!

Sean Pitman

unread,
May 6, 2004, 4:49:52 PM5/6/04
to
"Zachriel" <sp...@zachriel.com> wrote in message news:<_sydnWc3S7X...@adelphia.com>...

> > > The number of mutants is on the order of L^3 where L is the length of
> > > the string.
> >
> > Here is where you make one of your mistakes in conceptualizing the
> > problem. The total number of possible mutants is C^L where C is the
> > number of available characters and L is the length of the string.
> > That IS the total number of possible mutants for a character string of
> > a certain length. There just is no way around this concept.
>
> That is incorrect. The rules for mutation were clearly defined. Do you now
> want to add additional or different rules of mutation? We have
> point-mutations, delete mutations, insert mutations, snips, remainers and
> recombinations.

You just don't seem to get it. It doesn't matter how many kinds of
mutations are involved. A sequence of a given length can only have
C^L different arrangements. Your L^3 number is completely meaningless
in that it says nothing about how many "meaningful" sequences are
likely to be clustered around your starting point or the average time
it takes to find a new meaningful sequence.

> "Just try a little experiment yourself. Start with a short 2 or 3-letter
> word and see how many words you can evolve . . . "
> http://tinyurl.com/33otq

Exactly! Start with a short 2- or 3-letter meaningful sequence. Or,
if you want more than a steady state population of 1, start, start
with many such sequences - say 100 of them. Now, with a reproductive
rate of 2 per sequence per generation and an average mutation rate per
sequence that is 1 or greater, the maximum number of new sequences
that can be subjected to selection per generation is nothing more than
the number of offspring (O) per generation. In this case, the number
of offspring would be 100 * 2 or 200. So, a maximum of only 200 novel
sequences could be analyzed per generation. This is regardless of the
length of the sequences in the population (except for 1-letter only
sequences in this particular case).

Obviously then, your L^3 calculation is completely useless. I have no
idea what you think it tells you. It says nothing about the average
time needed to find a new much less meaningful sequence nor does it
accurately predict the density of meaningful sequences around a given
starting point.

> Please tell us, how many valid mutations can be made to this word, "at"? How
> many of these make valid words?

The answer to your first question is that it depends upon your
population size, reproduction rate, and mutation rate.

For example, if you have a steady state population of, say, 10
individual "at" sequences in your population and a reproductive rate
of 10 per individual "genome" per generation, then, in the first
generation, you could produce 100 offspring sequences that would be
subject to "selection". With a mutation rate of at least one mutation
per genome per generation, the odds that any two of these 100
offspring would be the same 2-letter sequence are fairly low (around 1
in 6 tries). That means that in most generations, all of the 100
offspring will be different. Now, the question is, out of these 100
offspring, how many of them will be uniquely "meaningful" in the
English language system?

Well now, that also depends now doesn't it? The answer to this
question is really the answer to your second question. Technically
speaking, the English language system _could_ have been set up so that
all 2-letter sequences surrounding the "at" sequence would be
meaningfully defined. If true, all 100 offspring sequences would be
meaningful. The problem is that English did _not_ happen to cluster
meaningful sequences so tightly together in one small corner of the
available sequence space. As it turns out, the meaningful 1-, 2-, and
3-letter sequences are fairly spread out in sequence space relative to
the starting "at" sequence. Of course, they are still all pretty much
linked together via very common single-mutation bridges all over the
place. But, the due to the spread out nature of the English language
system, only a relatively small fraction (about 1 in 5 or so at this
level) of our 100 offspring sequences (about 20 total), will be
meaningfully selectable using the English language system.

But, even though 20 offspring are positively "selectable" out of the
100, only 10 can be chosen to populate the next generation. Getting
rid of the 80 meaningless sequences is easy, based on a lack of
selectable advantage as compared with their peers. But, how is the
field of 20 selectable offspring narrowed down to the required 10
needed to maintain a "steady state" population? Well, since some of
these 20 may be 3-characters long, these would be preferentially
selected over shorter meaningful 1- and 2-character genomes. But, say
there are only five 3-letter character strings out of the 20. This
leaves 5 spots opening positions for the next generation with 15
competing offspring. Of these, lets say that 3 of them are single
letter meaningful sequences. These are eliminated in favor of the 13
higher scoring 2-letter meaningful sequences. Still though, only 5
spots are available. How are the 5 chosen out of the 13 equally
qualified offspring applicants? Well, this must be done by random
selection since there is no selective advantage of any one of the 13
two-letter sequences over any other.

Now, we begin our second generation with a brand new population of 10
sequences made up of five 2-letter and five 3-letter sequences for a
total of 10 genomes. Odds are quite good that no two sequences in
this second generation population will be the same. So, each one of
these 10 unique genomes produces "offspring" sequences for the next
generation, just like the first generation did. How many different
offspring will be produced in the second generation? The answer is
exactly the same. The total number of offspring in the second
generation will also be 100. Of these, odds are pretty good that they
too will all be unique . . . and so on for each generation.

In this manner, the maximum amount of sequence space searched over an
extended course of time is determined by multiplying the average
number of offspring per generation with the number of generations,
given that the mutation rate is at least 1 per genome per generation.
If the mutation rate were lower, the possible search space would be
proportionately smaller over the same period of time.

So you see, your "L^3" number doesn't say anything meaningful at all.
For example, a 2-character sequence, as in this illustration of yours,
would have a "Zach Number" of 2^3 or 8 - right? What the heck does
the number "8" tell us in this situation Zach? What is it supposed to
represent? Please Zach, do explain . . .

Sean
www.naturalselection.0catch.com

Zachriel

unread,
May 6, 2004, 9:53:48 PM5/6/04
to

"Sean Pitman" <seanpi...@naturalselection.0catch.com> wrote in message
news:80d0c26f.04050...@posting.google.com...
> "Zachriel" <sp...@zachriel.com> wrote in message
news:<_sydnWc3S7X...@adelphia.com>...
>
> > > > The number of mutants is on the order of L^3 where L is the length
of
> > > > the string.
> > >
> > > Here is where you make one of your mistakes in conceptualizing the
> > > problem. The total number of possible mutants is C^L where C is the
> > > number of available characters and L is the length of the string.
> > > That IS the total number of possible mutants for a character string of
> > > a certain length. There just is no way around this concept.
> >
> > That is incorrect. The rules for mutation were clearly defined. Do you
now
> > want to add additional or different rules of mutation? We have
> > point-mutations, delete mutations, insert mutations, snips, remainers
and
> > recombinations.
>
> You just don't seem to get it. It doesn't matter how many kinds of
> mutations are involved.

Sure it does. According to Dr. Pitman, "What you fail to do is to take into
account all the other possible arrangements and potentially non-beneficial
connections and insertions of these words."

So apparently, in the opinion of Dr. Pitman, it is important. Indeed, I
built two machines; one checks every "possible arrangement and potentially
non-beneficial connections and insertions", while the other tries random
"arrangements and potentially non-beneficial connections and insertions."


> A sequence of a given length can only have
> C^L different arrangements. Your L^3 number is completely meaningless
> in that it says nothing about how many "meaningful" sequences are
> likely to be clustered around your starting point or the average time
> it takes to find a new meaningful sequence.
>
> > "Just try a little experiment yourself. Start with a short 2 or 3-letter
> > word and see how many words you can evolve . . . "
> > http://tinyurl.com/33otq
>
> Exactly! Start with a short 2- or 3-letter meaningful sequence. Or,
> if you want more than a steady state population of 1, start, start
> with many such sequences - say 100 of them.

Please note that the challenge was to start with one.


> Now, with a reproductive
> rate of 2 per sequence per generation and an average mutation rate per
> sequence that is 1 or greater, the maximum number of new sequences
> that can be subjected to selection per generation is nothing more than
> the number of offspring (O) per generation. In this case, the number
> of offspring would be 100 * 2 or 200. So, a maximum of only 200 novel
> sequences could be analyzed per generation.

Well, the Word Mutagenator only analyzes one (random) mutation at a time, as
soon as they occur. It still evolves long words in much less than Pitman
Time (zillions of years).


> This is regardless of the
> length of the sequences in the population (except for 1-letter only
> sequences in this particular case).
>
> Obviously then, your L^3 calculation is completely useless. I have no
> idea what you think it tells you.

My goodness Sean. This is exactly what you insisted I calculate.

"What you fail to do is to take into account all the other possible
arrangements and potentially non-beneficial connections and insertions
of these words."
http://tinyurl.com/2te3t

I took them into account two different ways. The Word Mutator assumes a
large population, considers that each possible mutation will happen in each
generation, and that the generations were in breeding seasons; that is, all
the viable young would enter the population simultaneously. *This was
according to the posted rules.*

The other way though, the Word Mutagenator, treats each (random) mutation as
soon as it occurs. There is not distinct breeding season. And yet both
programs have very similar outcomes.

> It says nothing about the average
> time needed to find a new much less meaningful sequence nor does it
> accurately predict the density of meaningful sequences around a given
> starting point.

Who cares? The Word Mutagenator evolves new and longer words by simple
random mutation and mechanical selection for length.


> > Please tell us, how many valid mutations can be made to this word, "at"?
How
> > many of these make valid words?
>
> The answer to your first question is that it depends upon your
> population size, reproduction rate, and mutation rate.

This is false. According to everything you have stated previously on this
problem, we change one letter at a time, and see how far we can go. You have
moved the goalposts.


> For example, if you have a steady state population of, say, 10
> individual "at" sequences in your population and a reproductive rate
> of 10 per individual "genome" per generation, then, in the first
> generation, you could produce 100 offspring sequences that would be
> subject to "selection". With a mutation rate of at least one mutation
> per genome per generation, the odds that any two of these 100
> offspring would be the same 2-letter sequence are fairly low (around 1
> in 6 tries). That means that in most generations, all of the 100
> offspring will be different. Now, the question is, out of these 100
> offspring, how many of them will be uniquely "meaningful" in the
> English language system?

The Word Mutagenator uses random mutation. There is a population of "sean"
living happily in a Pond. One day, one of the "sean" mutates into "bean".
Now there are two species of words living in the Pond.


> Well now, that also depends now doesn't it? The answer to this
> question is really the answer to your second question. Technically
> speaking, the English language system _could_ have been set up so that
> all 2-letter sequences surrounding the "at" sequence would be
> meaningfully defined.

It wasn't.


<snip>


> In this manner, the maximum amount of sequence space searched over an
> extended course of time is determined by multiplying the average
> number of offspring per generation with the number of generations,
> given that the mutation rate is at least 1 per genome per generation.
> If the mutation rate were lower, the possible search space would be
> proportionately smaller over the same period of time.

And yet the Word Mutagenator works exactly according to your analogy and
despite your disbelief, finds long words in much less than a Pitman Zillion.


>
> So you see, your "L^3" number doesn't say anything meaningful at all.
> For example, a 2-character sequence, as in this illustration of yours,
> would have a "Zach Number" of 2^3 or 8 - right? What the heck does
> the number "8" tell us in this situation Zach? What is it supposed to
> represent? Please Zach, do explain . . .

You don't read well for comprehension. The L^3 figure is an upper-limit for
large L. This is also on the website. After so many words, and after
expressing so much confidence in your views, you should at least be familiar
with the argument.

The actual number of mutations of a 2-character sequence is 145.

Point Mutations 26*2 = 52
Insert Mutations 26*3 = 78
Snippets 2+1 = 3
Remainders 2+1 = 3
Snip-inserts = Snippets*3 = 9
Total = 145.

This number is a bit high due to some double-counting. Here are the numbers
for a variety of different L.
http://www.zachriel.com/mutagenation/calcs.xls


These are the valid mutations of "at" (according to the established rules).

Point mutations
aa ab ac ad ae af ag ah ai aj ak al am an ao ap aq ar as at au av aw ax ay
az
at bt ct dt et ft gt ht it jt kt lt mt nt ot pt qt rt st tt ut vt wt xt yt
zt

Insert Mutations
aat bat cat dat eat fat gat hat iat jat kat lat mat nat oat pat qat rat sat
tat uat vat wat xat yat zat
aat abt act adt aet aft agt aht ait ajt akt alt amt ant aot apt aqt art ast
att aut avt awt axt ayt azt
ata atb atc atd ate atf atg ath ati atj atk atl atm atn ato atp atq atr ats
att atu atv atw atx aty atz

Snippets
a t at

Remainders
t a null

Snip-inserts
aat aat ata
tat att att
atat aatt atat

I'm not sure there is any point in continuing this discussion. You are not
capable, or simply unwilling, to bend to weight of argument. You are set in
your opinions and no manner of facts will change your mind. That change will
have to come from within yourself. However, the results of Mutagenation
speak for themselves, and it is clear that very few will take your
assertions about "words" or genes seriously.

This was the challenge.

Just try a little experiment yourself. Start with a short 2 or 3-letter word

and see how many words you can evolve that require greater and greater


minimum sequence requirements. No doubt you will quickly find yourself

coming to walls of meaningless or non-beneficial potential options that
separate you from every other meaningful and beneficial option."
http://tinyurl.com/ypos7

Word Mutagenator can be started with a 2- or 3-letter word. It randomly
mutates them according to the simple rules everyone recognizes as mutation
(point mutation, recombination, etc.). It prunes their children for length.
It is done mechanically and without human intervention. The code is open
source, and the algorithm does not look ahead into the future. Yet long
words evolve in much less than a Pitman Time.

This was the claim.

"Getting from one meaningful 7-letter phrase to a different meaningful
7-letter phrase requires, on average, a fairly long random walk through
250,000 meaningless options."
http://tinyurl.com/ypos7

You were simply wrong. We don't have to walk through all that space. Get
over it.
http://www.zachriel.com/mutagenation/

Sean Pitman

unread,
May 7, 2004, 12:16:45 PM5/7/04
to
"Zachriel" <sp...@zachriel.com> wrote in message news:<HcidnSNerPH...@adelphia.com>...

> > You just don't seem to get it. It doesn't matter how many kinds of
> > mutations are involved.
>
> Sure it does. According to Dr. Pitman, "What you fail to do is to take into
> account all the other possible arrangements and potentially non-beneficial
> connections and insertions of these words."

The number of possible types or kinds of mutations does increase the
number of _possible_ mutation per mutation event. However, it does
NOT increase the number of mutations that can be _analyzed_ by the
selection process per generation. The number of analyzable mutations
is dependent upon the colony size, number of offspring, and average
_rate_, not kinds, of mutations.

> So apparently, in the opinion of Dr. Pitman, it is important. Indeed, I
> built two machines; one checks every "possible arrangement and potentially
> non-beneficial connections and insertions", while the other tries random
> "arrangements and potentially non-beneficial connections and insertions."

In a given population, you cannot "test" every possible arrangement
unless every possible arrangement actually evolves in your population
in every generation. The only way you could possibly do this beyond
very low levels is to have either an incredibly large population
replicating at a reasonable reproductive rate and having a reasonable
mutation rate, or a small population with an incredibly large
reproductive rate and an incredibly high mutation rate. You opted for
the second scenario.

If, on the other hand, you had programmed your computer to work like I
have suggested, a small steady state population of sequences with a
small reproductive rate and an average mutation rate of 1 per genome
per generation would have resulted in an exponential increase in the
average number of generations it would have taken your computer
population to reach _each_ higher level of sequence length complexity.



> > A sequence of a given length can only have
> > C^L different arrangements. Your L^3 number is completely meaningless
> > in that it says nothing about how many "meaningful" sequences are
> > likely to be clustered around your starting point or the average time
> > it takes to find a new meaningful sequence.

Right?

> > > "Just try a little experiment yourself. Start with a short 2 or 3-letter
> > > word and see how many words you can evolve . . . "
> > > http://tinyurl.com/33otq
> >
> > Exactly! Start with a short 2- or 3-letter meaningful sequence. Or,
> > if you want more than a steady state population of 1, start, start
> > with many such sequences - say 100 of them.
>
> Please note that the challenge was to start with one.

You're giving me a headache Zach! You certainly can start with a
population of 1, such as 1 bacterium in a petri dish. But the
starting point doesn't have to stay at one since the "1" can quickly
turn into "the many". What matters is the limitations of your
environment to the maximum size of the population - or the maximum
steady state of the population. Certainly you could set your maximum
steady state to 1, but then your enormous reproductive rate of many
thousands "per generation" in your simulation is required to make it
look like real life organisms can also evolve at equivalent rates -
and they just can't since even the most prolific bacterial
reproductive rate falls far short of your simulation. In my challenge
to you, my calculations where at least based on an attempt to simulate
real life limitations to at least a vaguely relevant degree. Your
computer simulation doesn't even come close. And, even what it does
do has not and cannot go beyond very low levels of relative English
language complexity, such as a meaningful much less beneficial work
requiring just a few hundred to a thousand or so characters, in a
practical eternity of time.

> > Now, with a reproductive


> > rate of 2 per sequence per generation and an average mutation rate per
> > sequence that is 1 or greater, the maximum number of new sequences
> > that can be subjected to selection per generation is nothing more than
> > the number of offspring (O) per generation. In this case, the number
> > of offspring would be 100 * 2 or 200. So, a maximum of only 200 novel
> > sequences could be analyzed per generation.
>
> Well, the Word Mutagenator only analyzes one (random) mutation at a time, as
> soon as they occur.

The real life equivalent to this would be a bacterial colony with a
steady state of 1. Each time a bacterium mutates, it is instantly
subject to natural selection and each of these mutations will be
passed on to the offspring of that bacterium.

This is different from mutations in creatures that use sex and genetic
recombination for reproduction. The only mutations that count in
these creatures are those that affect the germ cell line. But, germ
cell mutations are not subject to selection pressures until they are
expressed in the next generation.

Your program actually works the second way since you put all mutants
in a collective "basket" before subjecting them to 'batch selection"
in the "next generation". This means that your computer program does
not analyze each random mutation as soon as it occurs. It batches
them in a pool of mutants and then analyzes them in bulk. The only
way you could do this as a model of real life, is to call each
separate mutant that your computer will batch analyze a separate
offspring.

Now, if you actually did analyze each mutation as it occurred, as in
the bacteria-type scenario, you must discard the previous sequence
with each mutation in a steady-state population of 1. Now, if you had
a larger population all starting with the same sequence and made up of
say, 10 genomes, your population as a whole could maintain a
particular sequence _if_ the reproductive rate was high enough
relative to the mutation rate. For example, all 10 genomes in your
starting population could all read, "start". You suggest that all of
these 10 start sequences just wait around in the population pool
waiting to mutate and then, when mutated, they are instantly selected
for or against survival - right? Well, say a random mutation affects
one of the individuals in your population. What are the odds that
this one mutation will be uniquely meaningful and therefore positively
selectable? Not very good right? - and worse at each higher level.
If this mutation were detrimental in that it destroyed the meaning of
the previous sequence, what happens to that individual in your
population? It is deselected and "dies out" of your population. Now,
your population is smaller. How is the size of your population
maintained? Obviously the genomes in your population must replicate
by producing copies of themselves or "offspring" - right?

Say the reproductive rate is also 10 offspring per second per genome
and the mutation rate averages 1 mutation per genome per reproductive
event. In the first second, how many offspring would be made?
Obviously the answer averages 100 - right? Considering the
parameters, how many of the 100 offspring would have the same
sequence? On average, none of them would have the same sequence.
There would be 100 different offspring. Now, each of these offspring
as well as their parents are analyzed continuously and either kept or
discarded by a relative comparison to the other members of the
population to see if one genome is more or less fit than its
companions. The computer "environment" continuously selects just
enough to maintain the population at a constant steady state of 10
over the course of time. All of the rest of the genomes are
eliminated from the population.

So how many mutations are analyzed on average during each second of
time? 100 - right? Now, if you raise the mutation rate to 10
mutations per genome per reproductive event, how many mutations can be
analyzed per second? It may surprise you that in such a situation the
answer would still be 100 since mutations created during a
reproductive event are not selectable until the reproduction is
complete - as in your "batch" selection process. So, over the course
of 100 seconds, how many different sequences could be analyzed on
average? 10,000 - right?

Of course, your scenario is basically the same except that you use a
much much higher reproductive rate, mutation rate, and even
population. Of course, raising these parameters makes evolution
possible at higher levels over the course of a given amount of time.
The problem is that these parameters have to be raised in an
exponential manner to keep up with a linear increase in the goal-level
of complexity. This means that even a computer environment and
population that is capable of incredibly high reproductive rates,
mutation rates, and even population sizes, still stalls out at
relatively low levels of meaningful sequence complexity (i.e., less
than a few hundred to a thousand or so fairly specified
meaningful/beneficial sequence characters in a language/information
system).


> It still evolves long words in much less than Pitman
> Time (zillions of years).

These are not "long" sequences at all - relatively speaking and the
time involved to evolve even what your computer did evolve was
dramatically reduced by the huge reproductive rate and mutation rate
programmed into your population. You must see that this is obviously
true - right?



> > This is regardless of the
> > length of the sequences in the population (except for 1-letter only
> > sequences in this particular case).
> >
> > Obviously then, your L^3 calculation is completely useless. I have no
> > idea what you think it tells you.
>
> My goodness Sean. This is exactly what you insisted I calculate.
>
> "What you fail to do is to take into account all the other possible
> arrangements and potentially non-beneficial connections and insertions
> of these words."
> http://tinyurl.com/2te3t

You use L^3 for "larger sequences" as an estimate of the number of
potential mutations from the perspective of a given string per
mutation event - correct? What you evidently fail to realize is that
this number of _potential_ mutations from the perspective of one
character string is not the same thing as the number of _potential_
mutations from the perspective of a population of strings. And, even
from the perspective of a population, the number of _potential_
mutations is not the same thing as the number of _possible_ mutations.
The number of potential mutations basically maintains the size of
sequence space, which you were seemingly trying to over look in your
initial scenarios. However, now you are trying to overlook the fact
that the number of possible analyzable mutations per generation in a
steady state population has nothing at all to do with the number of
potential mutations. The number of possible analyzable mutations for
a population in a given generation is NOT L^3, but is rather
determined by the number of offspring per generation TIMES the average
number of mutations per offspring per generation. That is the number
of possible analyzable mutations.

Do you understand the difference between potential and possible
mutations as I am using these terms in the current sense? There is a
difference and an understanding of this difference is important to
understanding this problem.

> I took them into account two different ways. The Word Mutator assumes a
> large population, considers that each possible mutation will happen in each
> generation, and that the generations were in breeding seasons; that is, all
> the viable young would enter the population simultaneously. *This was
> according to the posted rules.*

Note also that your argument assumes that all sequences in your "large
population" are all the same in each generation - which is simply not
true. In each of your generations your population of sequences
changes dramatically and they are all different. This translates into
a huge reproductive rate since each _potential_ mutation happens in
each "generation". As previously discussed, in order to have a
selectable mutation in your generation, you must have an individual
sequence "offspring" that represents that sequence. This means that
if every potential sequence is produced and each of these sequences
must be represented by a different offspring, that a huge number of
offspring must be produced as well along with a very large random
mutation rate (which is not exactly random in this setup).

> The other way though, the Word Mutagenator, treats each (random) mutation as
> soon as it occurs. There is not distinct breeding season. And yet both
> programs have very similar outcomes.

This is because both programs use extremely high reproductive rates
and mutation rates, as previously described. And yet, your "outcomes"
do not go very high up the ladder of complexity, relatively speaking.
Your outcomes are still significantly limited to relatively low levels
of functional sequence complexity.



> > It says nothing about the average
> > time needed to find a new much less meaningful sequence nor does it
> > accurately predict the density of meaningful sequences around a given
> > starting point.
>
> Who cares?

Who cares? I care and so should you because your L^3 calculation is
completely irrelevant to the problem if it doesn't say or predict
anything about the actual workings or outcome of the problem. Your
outcome is not based at all on the L^3 calculation, but on the very
high mutation rates and reproductive rates in a population that is
working at very low relative levels of meaningful sequence complexity.
Your extrapolations of your L^3 formula to much longer sequences than
you actually were able to "evolve" with your computer are therefore
just as irrelevant. Your conclusions that much more complexity
meaningful sequences could be evolved in much less than even "millions
of years" are therefore based on an irrelevant calculation and are
therefore meaningless.

> The Word Mutagenator evolves new and longer words by simple
> random mutation and mechanical selection for length.

Again, your outcome is not based at all on the L^3 calculation like
you claim, but on the very high mutation rates and reproductive rates
in a population that is working at very low relative levels of
meaningful sequence complexity. Your extrapolations are therefore just
as irrelevant.



> > The answer to your first question is that it depends upon your
> > population size, reproduction rate, and mutation rate.
>
> This is false. According to everything you have stated previously on this
> problem, we change one letter at a time, and see how far we can go. You have
> moved the goalposts.

I haven't moved the "goalposts" at all. It is just that your
understanding of the problem has been warped from the beginning and is
still at a very elementary level. Obviously "changing one letter at a
time" in a population of sequences that reproduces and mutates in a
generation-type way, is very much affected by the population size,
reproduction rate, and mutation rate. If you can't understand or
agree with this, you are simply beyond me.



> > For example, if you have a steady state population of, say, 10
> > individual "at" sequences in your population and a reproductive rate
> > of 10 per individual "genome" per generation, then, in the first
> > generation, you could produce 100 offspring sequences that would be
> > subject to "selection". With a mutation rate of at least one mutation
> > per genome per generation, the odds that any two of these 100
> > offspring would be the same 2-letter sequence are fairly low (around 1
> > in 6 tries). That means that in most generations, all of the 100
> > offspring will be different. Now, the question is, out of these 100
> > offspring, how many of them will be uniquely "meaningful" in the
> > English language system?
>
> The Word Mutagenator uses random mutation. There is a population of "sean"
> living happily in a Pond. One day, one of the "sean" mutates into "bean".
> Now there are two species of words living in the Pond.

The pond is still made up of a limited "population" of sequences. You
just have two different kinds of sequences now in your pond. Yet,
this does not change the fact that the population size, reproductive
rate, and mutation rate does significantly effect the rate of
evolution of your population as a whole - as described in detail
several times before.



> > Well now, that also depends now doesn't it? The answer to this
> > question is really the answer to your second question. Technically
> > speaking, the English language system _could_ have been set up so that
> > all 2-letter sequences surrounding the "at" sequence would be
> > meaningfully defined.
>
> It wasn't.

That is correct. It wasn't set up like this even though it could have
been. Instead, it was set up very much like I claim it was. It is
much more randomly diffuse in its setup than you and many other
evolutionists seem to be capable of recognizing. At lower levels the
islands and bridges are in fact quite common. But, as even you have
discovered, these islands start moving rapidly away from each other
and the bridges start narrowing and snapping completely, in an
exponential manner, with each step up the ladder of meaningful
complexity.



> <snip>
> > In this manner, the maximum amount of sequence space searched over an
> > extended course of time is determined by multiplying the average
> > number of offspring per generation with the number of generations,
> > given that the mutation rate is at least 1 per genome per generation.
> > If the mutation rate were lower, the possible search space would be
> > proportionately smaller over the same period of time.
>
> And yet the Word Mutagenator works exactly according to your analogy and
> despite your disbelief, finds long words in much less than a Pitman Zillion.

Yes, it does work according to my analogy, but not despite my
disbelief. I have explained over and over to you why your
"Mutagenator" works _according_ to my belief. In fact, why do you
think I was not more skeptical of your findings. As soon as you
declared your outcome, I immediately accepted it without any problem
or balking. Now why is that do you think? It is because your outcome,
given the way in which you set up your program, was perfectly
predictable. However, your extrapolations to very long meaningful
sequence evolution, based on your very low level outcome, are not
statistically possible at all. For example, nothing equivalent to
your "O Sean Pitman" poem will ever evolve, even with the very high
reproductive rates and mutation rates programmed into your computer.
The average time required simply is not a function of L^3 since L^3
says nothing about the location of meaningful sequences in sequence
space . . . and that is was really determines the average time
required.

> > So you see, your "L^3" number doesn't say anything meaningful at all.
> > For example, a 2-character sequence, as in this illustration of yours,
> > would have a "Zach Number" of 2^3 or 8 - right? What the heck does
> > the number "8" tell us in this situation Zach? What is it supposed to
> > represent? Please Zach, do explain . . .
>
> You don't read well for comprehension.

LOL - ok . . .

> The L^3 figure is an upper-limit for
> large L. This is also on the website. After so many words, and after
> expressing so much confidence in your views, you should at least be familiar
> with the argument.

Oh, I am. But the L^3, even for a large L, says nothing about what a
population is capable of nor does it say if anything within the L^3,
from the perspective of a single sequence string, is in fact
meaningful or what the odds are that there might even be something
meaningful within the L^3 distance. As previously described, the
meaningful sequences in sequence space could have been set up so that
all sequences within the L^3 distance were meaningfully beneficial or
that non of these sequences were meaningful much less beneficial. Your
L^3 does not answer or even remotely estimate the possibility or
probability of either of these options or of the continuum that exists
between these options for any level of complexity.

Are you starting to see how completely worthless your stated
extrapolations of what your L^3 calculation really are? Your
extrapolations are based on a meaningless calculation and are
therefore just as meaningless and irrelevant to the actual problem at
hand.

> The actual number of mutations of a 2-character sequence is 145.
>
> Point Mutations 26*2 = 52
> Insert Mutations 26*3 = 78
> Snippets 2+1 = 3
> Remainders 2+1 = 3
> Snip-inserts = Snippets*3 = 9
> Total = 145.
>
> This number is a bit high due to some double-counting. Here are the numbers
> for a variety of different L.
> http://www.zachriel.com/mutagenation/calcs.xls

Again, this 145 number, even if entirely correct, says absolutely
nothing about how many of these 145 sequences are selectable as
"meaningful/beneficial". And yet, this is one of the most important
questions. Also, this 145 number says nothing about the other members
of a population. This means that it says nothing about the total
number of selectable mutations that a population can realize during a
mutation "event". The population number would be somewhere around 145
times the number of individuals in the population. Depending upon the
population size and mutation rate, this number could easily be as high
as the total 2-letter sequence space itself - or 676 different
sequences.

Your calculations simply do not deal with or even recognize this very
real and very important possibility. You are basically blind to this
possibility and so this leads to your erroneous extrapolations of your
L^3 calculation.

> I'm not sure there is any point in continuing this discussion. You are not
> capable, or simply unwilling, to bend to weight of argument.

LOL - Look Who's Talkin!

Now, I'm quite sure that you think you are just as clear in your
thinking as I think that I am clear in my thinking. However, to be
honest with you, as far as evolutionists that I have seriously
debated, you certainly don't come across as the brightest bulb in the
box. At least those like Ian Musgrave, and few others, at least come
across as both intelligent and informed on the issues.

> You are set in
> your opinions and no manner of facts will change your mind.

That certainly does seem to be your problem as well. Despite the fact
that even you recognize that my suggested evolutionary scenarios are
"very good", you don't seem to understand why they are good.

> That change will
> have to come from within yourself. However, the results of Mutagenation
> speak for themselves,

Yes, they certainly do for anyone who really understands what you are
actually doing with your model.

> and it is clear that very few will take your
> assertions about "words" or genes seriously.

Unfortunately this may be true. But, as a general rule, it is the few
and not the many who come up with and initially even understand the
really good ideas.

Sean
www.naturalselection.0catch.com

RobinGoodfellow

unread,
May 7, 2004, 3:34:14 PM5/7/04
to
Sean Pitman wrote:

> "Zachriel" <sp...@zachriel.com> wrote in message news:<HcidnSNerPH...@adelphia.com>...
>

[snip]

>>and it is clear that very few will take your
>>assertions about "words" or genes seriously.
>
>
> Unfortunately this may be true. But, as a general rule, it is the few
> and not the many who come up with and initially even understand the
> really good ideas.
>
> Sean
> www.naturalselection.0catch.com
>

Ah, yes, the last refuge of virtually every kook, and the odd visionary
who comes up with the occasional good idea, usually long before its time
(i.e. before it can be reasonably tested). Sorry, Sean, but you'll have
to forgive everyone here if we don't recognize you for the visionary
that you are. So far, people with far better training than you in the
many disparate areas where you attempt to promote your ideas have all
agreed on one thing - your arguments are simplistic, poorly informed,
and thoroughly unconvincing, if not outright wrong. Now, it could be,
of course, that we are all blinded by our naturalistic, old-earth,
evolutinary presuppositions - even the theists among us (I happen to be
one). Or maybe none of us is bright enough to see the subtle brilliance
that lurks behind your simple arguments from biology and elementary
math. Perhaps it is a combination of the above that afflicts us all,
along with the overwhelming portion of biologists, chemists, physicists,
mathematicians, etc. who have spent their careers delving deep into the
problems you seem to have solved with a wave of your hand, and offerring
up answers which, in part or in whole, almost always contradict yours.
It is certainly possible. But it is also possible that Elvis is still
alive. Off the top of my head, I honestly can't say which possibility
appears to be more likely.

As a last-ditch effort, I started a new thread recently, where I ask you
for a more detailed explanation of your position. I want to make sure
that, for my part, I'm really not misunderstanding you, as you seem to
think we all do. If you get the chance, I hope you can answer some of
the questions I asked in there. I'd be happy to answer any of yours, in
case you have any. See the URL,

http://groups.google.com/groups?q=sean+pitman+group:talk.origins+robingoodfellow&hl=en&lr=&ie=UTF-8&oe=UTF-8&group=talk.origins&c2coff=1&selm=81fa9bf3.0405052256.4474fdff%40posting.google.com&rnum=1

or simply search TO for the topic "Sean Pitman, Sequence Space, and the
Complexity of Evolution".

I wish you luck in spreading your "really good ideas". Should your day
ever come, I'll gladly eat my words and get you Elvis's autograph.

Cheers,
RobinGoodfellow.

Sean Pitman

unread,
May 7, 2004, 4:39:19 PM5/7/04
to
"Zachriel" <sp...@zachriel.com> wrote in message news:<UbGdncC6usE...@adelphia.com>...

> > Oh, and by the way, by meaning I am talking about interactive meaning
> > where the sequence as a whole is greater in meaning than the sum of
> > its parts.
>
> And exactly how do you suppose we can program a computer to select for that?
> So you have indeed moved the goal-posts at supersonic speeds.

I haven't moved the goalposts at all since this is exactly where
anyone who wants to accurately model Darwinian evolution must place
their goalposts.

> > Again, you are looking at just the mutations available for one
> > individual here with a mutation rate of one and only one per
> > individual. If you look at the mutations available for the population
> > at large you will notice that there are far more possible mutations
> > per generation than you have loudly asserted on your website.
>
> Another specific claim. Time to put up, Sean. Please put "pitman" or
> whatever you want into the Word Mutator and find a mutant--according to the
> posted rules--that is not accounted for. Don't make claims for which you
> have no support. The code is open source. The possible mutations are clearly
> defined.

What claim, exactly, do you think I just made in the above paragraph
Zach? This response seems to me to be completely irrelevant - though
not entirely surprising. Again, you don't seem to be considering the
mutation potential of the population at large over the course of time,
but are only looking at the population from the perspective of a
single type of sequence over the course of a single mutation event.

> For a starting population of "sean" and "pitman" there are 31 possible valid
> words out of at most 1014 mutants.

You see what I'm saying here Zach. You only look at the possible
mutants for a single generation that only has one or two types of
seqeunces to start out with. You also fail to consider that a
population with a limited reproduction and/or mutation rate may not be
able to cover all of these "possibilities" in one generation. More
importantly, you fail to realize that in subsequent generations the
population will significantly diverge with each member of a population
potentially and even likely having a different sequence. This rapidly
and greatly increases the number of different mutants that may be
subjected to selection in each generation.

Also, you seem to fail to realize that your _ratio_ of 31 in 1014
possible mutants (about 1 in 32) for your starting 2-sequences in your
initial population, is much higher than it would be if you started
with smaller 2- or 3-letter words and much lower than if you started
with 9- or 10-letter words. This changing ratio shows an exponential
decline with increasing sequence length. Your L^3 calculation says
nothing about this declining ratio, now does it?

The fact of the matter is, there is nothing inherent about the English
language system or any other language system that requires this ratio
to decline like it does. All the meaningful sequences could have been
stacked around your starting point so that all 1014 possibilities were
meaningful. The fact that only 1 in 32 of them are meaningful is
pretty much arbitrary - wouldn't you say? The ratio could have been 1
in 1014 just as easily now couldn't? Or, it could have even been 0 in
1014. That fact that it is 1 in 32 is just the way it is, but it is
not the way it had to be - right?

So, what's the point? The point is that your L^3 calculation carries
with it absolutely no predictive value as to what the ratio will be at
any level - period. You therefore cannot predict how long meaningful
evolution will take at higher and higher levels based on this formula.

Your computer programs are quite successful at the very low levels
within which they work, which is not surprising at all especially
giving the parameters you chose (i.e., extremely high reproductive
rates and mutation rates, etc.). The problem is that I know why your
programs were successful while you don't seem to have the first clue.
You go off about some L^3 calculation that is completely irrelevant to
why your programs worked or why they would soon stall out with a
relatively few more steps up the ladder of meaningful complexity
(i.e., far less than the 1,000-character level). It seems that you
still just don't understand that your L^3 calculation means absolutely
nothing as far as predicting the rate of meaningful evolution in a
population over time - nothing. Please do and at least try really hard
to grasp this concept. It is not that hard. Really it isn't.

Sean
www.naturalselection.0catch.com

Sean Pitman

unread,
May 8, 2004, 10:03:37 AM5/8/04
to
RobinGoodfellow <lmuc...@yahoo.com> wrote in message news:<c7gom4$1no$1...@news01.cit.cornell.edu>...

This is a classic evolutionists comeback. How can a YEC possibly be
right when the whole world of brilliance stands against the poor
deluded YEC? The question is though, if those who stand against me
are so brilliant and understand how evolution works so well, why is it
so difficult for them to explain where I am wrong? Of course you will
say that it is not difficult and they have done a fine job, but
clearly they haven't. Your own arguments illustrate this point very
nicely. In order to attempt to answer the neutral gap problem, you
simply come up with the "stacked deck" idea, which, although it would
solve the problem of evolution, it wouldn't solve the problem of how
the deck got so neatly stacked without intelligent design? Also, you
have shown absolutely no evidence that the deck is stacked as neatly
as you believe while I have shown a great deal of evidence that it is
not stacked and that what islands and bridges exist at lower levels
quickly break down, in an exponential manner, with each step up the
ladder of functional complexity until evolutionary potential
completely breaks down this side of a practical eternity of time.
Because of this exponential decline in evolutionary potential, the
existence of very high levels of specified complexity within all
living things shouts out "intelligent design" since truly mindless
evolutionary processes would simply require an practical eternity of
time to achieve such high-level systems.

In short, it is easy to say that someone's ideas are "crazy", but it
is quite another thing to logically and clearly explain why they are
crazy. You certainly haven't done much more than wild hand waving and
meaningless smoke blowing as far as I can tell. Your reasoning, your
Monte Carlo Math, and your attempts to intelligently "stack the deck"
are almost laughable as you try to apply them to this problem.

However, I will try yet again to look at your new thread and answer
your questions. Good luck in finding Elvis though! ; )

Sean
www.naturalselection.0catch.com

David Jensen

unread,
May 8, 2004, 10:24:25 AM5/8/04
to
In talk.origins, seanpi...@naturalselection.0catch.com (Sean Pitman)
wrote in <80d0c26f.04050...@posting.google.com>:

It appeared to me that many folks have done a fine job of explaining
where and why you are wrong. It appears that the problem comes at your
end.

Sean Pitman

unread,
May 8, 2004, 4:29:59 PM5/8/04
to
David Jensen <da...@dajensen-family.com> wrote in message news:<tlrp909vtsjlcunti...@4ax.com>...

Read the next line where I quote your very words ("fine job") before
you even said them. Is it me or am I starting to sound just a bit
prophetic?

David Jensen

unread,
May 8, 2004, 6:25:34 PM5/8/04
to

No, you just repeat yourself. You are the one who is showing that you
cannot or will not understand science.

0 new messages