Weasel program -- How evolutionists misrepresent ReMine
Walter ReMine
Ian Musgrave has reposted (from 1999) his misrepresentations about my book.
I responded to it at the time, and my posts from that era can still be
recovered from Google.com "Advanced Groups Search" by searching for the
author:
"laser...@my-deja.com".
For example, below is a copy (from 1999) of my first response to his
misrepresentations.
-- Walter ReMine
Fellow with Discovery Institute
Center for the Renewal of Science and Culture
_The Biotic Message_
http://www1.minn.net/~science
===================================================
===================================================
From: laser...@my-deja.com (laser...@my-deja.com)
Subject: Dawkins Simulation & Haldanes Dilemma
Newsgroups: talk.origins
Date: 1999/09/28
Dawkins Simulation & Haldane's Dilemma
Before I get to the technical details I must give some background.
Several years ago Wesley Elsberry and Clark Dorman began announcing
ill-formed opinions about my book, _The Biotic Message_. In many ways I
could tell they hadn't read it. (In the same way your old English
teacher could tell about your homework reading.) The compelling proof,
however, was that within months they were demanding to know about a
computer simulation regarding Haldane's Dilemma (that I had mentioned on
the Internet). They loudly demanded me to reveal the simulation, and
suggested I was hiding something. They kept up this empty posturing for
years, with me repeatedly answering, "It's all in the book." Until five
weeks ago, when another evolutionist posted that it actually is in my
book. Just as I said all along.
That began a round of recent posts on this subject. The simulation is
the "Methinks it is like a weasel" program discussed in Richard
Dawkins's book, _The Blind Watchmaker_. It is appropriately in my
chapter on Haldane's Dilemma. It takes an entire section (six pages),
and the name of the simulation is elevated to the name of section. You
cannot read the book and miss it. Quite simply, Elsberry and Dorman did
not bother to give my book an honest reading before they misrepresented
it for years.
Move now to Robert Williams, who for several years has run a website
that critiques my book's material on Haldane's Dilemma -- at least, that
is the impression his readers get. (His website now is virtually
identical to how it was several years ago.) Yet I could tell he hadn't
bothered to read my book, because he misrepresents my material numerous
times (both by commission and omission). So I pressed him about this in
a debate on sci.bio.evolution eighteen months ago, and he publicly
acknowledged he had not read my book -- the very material he was
pretending to critique on his website! (My publisher has since
contacted him numerous times asking him to clarify his website so his
readers may know this. He has refused.) Despite my public protests, his
website is still promoted, especially by the talk.origins crowd, as a
"critique" of my book. Yet Williams hadn't read my book.
Move forward again to a month ago. James Acker began posting
misrepresentations of my material, and I posted my objection. Acker then
publicly vowed to CONTINUE his [mis-]presentations even though he hasn't
read my book.
The remarkable thing, the important point, is that through-out all this
NOT ONE EVOLUTIONIST OBJECTED. Not one! Not one evolutionist said the
obvious, "Messrs. -- Elsberry, Dorman, Williams, and Acker -- you really
ought to give an honest reading to the material you are criticizing, at
a bare minimum, wouldn't that be the scholarly and honest thing to do?"
The community of Internet evolutionists holds a double standard -- they
demand scholarship from their opponents, but they do not demand it from
their fellows. That is the point I am documenting here today.
Now we move to the latest example of misrepresentation.
Ian Musgrave recently posted the following point. Due to the peculiar
inner construction of the Dawkins computer simulation, the reproduction
rate and population size are directly linked, unlike in nature. Thus,
when one lowers the reproduction rate by a factor of sixteen, that
lowers the population size by a factor of sixteen. Therefore he suggests
I had done something unrealistic to this simulation. Just who is he
kidding! Dawkins's simulation is unrealistic from the get-go -- wildly
in favor of evolution. No mere factor of sixteen changed that.
Yet Musgrave's post completely omitted that. He selectively quoted from
my book and SELECTIVELY OMITTED the vast bulk of its material revealing
the many unrealistic pro-evolution features of this simulation. There is
an abundance of that material, so abundant that much of it is neatly
stacked in densely packed, bulleted lists. It is impossible to read
that section of my book (six pages) and not be overwhelmed by its
presence. That material is directly relevant to the issue he raised --
Is the simulation unfair to evolution? My book is on top of his point.
It is hammering all over the point. Yet he omitted any mention of it!
He omitted any vague reference to it. He recklessly misrepresented my
book.
And once again, not one evolutionist objected. Not one. Not one said
the obvious, "Mr. Musgrave, I have seen ReMine's material, and you have
unfairly omitted the guts of his argument. Wouldn't it be better to
deal with his material as it actually is, rather than manufacture some
neutered version of it? Wouldn't it have been better to at least
mention ReMine's points that have a direct bearing on your issue?" No,
not one evolutionist objected to Musgrave's misrepresentation of my
work. On the contrary, Tim Ikeda (in a followup post) even found it
humorous.
I hope there will someday be an honest debate about origins (and my
material), but that is impossible in an environment that fosters and
embraces rampant misrepresentation. That is why thoughtful folks
routinely avoid such places, as I do.
Musgrave suggests it is 'unrealistic' to use a small population size in
the simulation. Doesn't that strike you as odd, since any other day of
the year evolutionists promote Dawkins's simulation to an unsuspecting
public. Let me use Musgrave's words to make the point:
"Trying to compare the substitution rate
in a population of [N=100] individuals
with the substitution rate in a population
of between 10,000 to 100,000 individuals
is a pretty big blunder to make"
(from Musgrave's post, brackets mine,
N=100 from Dawkins's simulation)
If Musgrave believes his own argument, then Dawkins made a "pretty big
blunder". If he wants to make charges, they must be placed on Dawkins
first. Followed by evolutionists as a whole, because they fully
promoted this simulation and failed to inform the public about its many
reckless pro-evolution features.
Musgrave had overlooked the key question: Just what about a small
population would DIS-favor the evolutionary simulation? This question
is obvious, yet not one evolutionist asked him about it. For them it was
enough merely that the simulation is "unrealistic". It didn't seem to
matter that in well over a dozen ways the simulation is unrealistic --
all wildly in FAVOR of evolution!
So let us focus on Musgrave's issue, population size. Dawkins's
simulation uses a small population size, which gives it a low cost of
substitution. Other things being equal, a low cost of substitution
means faster substitution. Given a species limited reproduction rate,
the population can be 'turned over' (or replaced with a new trait) much
quicker when the population is small. Population genetic change can
happen quickly in a small population. This is the advantage of a small
population, and this is in FAVOR of the simulated evolution.
Apply that to the simulation I ran for my book. Take Dawkins's
simulation. It uses an unrealistic deterministic mutation whose sole
purpose is to enhance the evolutionary illusion. Therefore correct it
to use a realistic method of random mutation. That slows the evolution.
Then reduce the reproduction rate (from N=100) to something a bit more
realistic (N=6), for a sexual species this would require the females to
produce 12 offspring each. The simulation then goes into error
catastrophe and never reaches the selection target. So we then give
evolution another favor by adjusting the mutation rate to eliminate the
error catastrophe and obtain the fastest speed of evolution. What
happens? The population is tiny -- its cost of substitution is also
tiny, at 5. And that tiny cost FAVORS the speed of evolution. It is
one-sixth the cost that Haldane estimated as the average for evolution
-- 30. This simulation incurs a tiny cost, and for a given reproduction
rate that makes for faster evolution. In Dawkins's simulation, the small
population size favors evolution. (And I have not even begun to recount
here the well over a dozen other ways this simulation favors evolution.)
Nonetheless, with all these advantages, this simulation is less than
five times faster than the rate known as Haldane's Dilemma.
Now my opponents ought consider the following question: For what reason
would a small population DIS-favor the evolutionary simulation? If you
try to find such a reason, you will fail. You will find the usual
reasons are neatly disabled by the design of Dawkins's simulation. His
simulation disallows the usual disadvantages of a small population. I
claim that this small population (with N=6) in fact has the advantages
of a large population -- without the high substitution cost of a large
population. Put simply, Dawkins's simulation is peculiarly designed to
have the advantages of small and large populations *at the same time*.
It has the best of both! A natural population cannot possibly do that,
but the simulation is unrealistically designed to do that. And it is all
in FAVOR of evolution.
Take it as a challenge. Think of some disadvantage of a small
population, then carry it through to see how Dawkins's design gets
around that disadvantage. After you've had this "Aha!" experience, then
notice that the reasons are clearly called out in my section on this
simulation, and explained in my book, and that Musgrave unscrupulously
omitted any mention of them.
Of all the evolutionary simulations, the one by Dawkins is the most
widely known. It rightly deserves to be debunked, and my book did so.
In the process it demonstrates the seldom-told principles discussed in
my chapter on Haldane's Dilemma.
-- Walter ReMine
_The Biotic Message_
http://www1.minn.net/~science
David Sienkiewicz
> Ian Musgrave has reposted (from 1999) his misrepresentations about my book.
> I responded to it at the time, and my posts from that era can still be
> recovered from Google.com "Advanced Groups Search" by searching for the
> author:
> "laser...@my-deja.com".
Indeed they can, as can the replies and resulting threads.
No one misrepresented you, Walter, and I saved $50 NOT buying your
book for my daughter, who is a biology student (now at the graduate
level).
For that, I thank the folks who reviewed and refuted you.
< snip >
Ian Musgrave & Peta O'Donohue
Address altered to avoid spam, delete RemoveInsert
On Sat, 10 Aug 2002 17:04:36 +0000 (UTC), Walter ReMine
<sci...@nospam.net> wrote:
>Ian Musgrave has reposted (from 1999) his misrepresentations about my book.
No, I've posted a web address of an updated version, that makes the
concept of Haldane's dilemma a little clearer (IMHO), and why ReMines
argument is profoundly mistaken.
http://www.health.adelaide.edu.au/Pharm/Musgrave/essays/remine.htm
>I responded to it at the time, and my posts from that era can still be
>recovered from Google.com "Advanced Groups Search" by searching for the
>author:
> "laser...@my-deja.com".
Or better musgrave AND weasel
>For example, below is a copy (from 1999) of my first response to his
>misrepresentations.
[big snip of stuff completely irrelevant to the issue at hand, then
and now]
>Musgrave suggests it is 'unrealistic' to use a small population size in
>the simulation.
Not unrealistic, downright incorrect (see below)
>Doesn't that strike you as odd, since any other day of
>the year evolutionists promote Dawkins's simulation to an unsuspecting
>public. Let me use Musgrave's words to make the point:
>
> "Trying to compare the substitution rate
> in a population of [N=100] individuals
> with the substitution rate in a population
> of between 10,000 to 100,000 individuals
> is a pretty big blunder to make"
> (from Musgrave's post, brackets mine,
> N=100 from Dawkins's simulation)
Yes, but then they are NOT my words when you do that, (check to
original for the real usage)
>If Musgrave believes his own argument, then Dawkins made a "pretty big
>blunder". If he wants to make charges, they must be placed on Dawkins
>first. Followed by evolutionists as a whole, because they fully
>promoted this simulation and failed to inform the public about its many
>reckless pro-evolution features.
Dawkins didn't make any blunder, as Dawkins was never making any claim
about Haldane's dilemma. Nor is dawkins program a simulation of
natural selection.
Haldane's "dilemma" is that, with certain assumptions, and in certain
broad ranges, it will take just as long to "fix" (ie go from being
very rare to being present in all members of the population) two
simultaneous beneficial mutations (ie the organism have two benefical
mutations in the same organism) in a population as it takes to fix two
sequential mutations (ie that it takes longer to fix a beneficial
double mutant than a beneficial single mutant see R Williams page in
the links section of my page for mathematical discussion of why this
should be so).
Mr. ReMine claims that Dawkins weasel program demonstrates Haldane's
dilemma. Yet in Dawkins program (and the majority of those programs
that replicate dawkins original), beneficial mutations are
automatically fixed in a _constant_ 1 generation. Dawkins program (and
Dawkins-style programs, such as on my page) CANNOT demonstrate
anything about Haldane's dilemma.
Dawkins program was not designed to be a simulation of natural
selection. Dawkins says this explicitly and mentions some of the
reasons that it isn't, and what a real simulation would entail.
What it _is_ is a demonstration of the efficacy of selection over
random search (which it does quite admirably). A random sequence of
letters is copied several times, and (some or all, depending on the
version of the program) of the copies are mutated. A single "best"
copy (determined by an algorithm that compares all the copies to a
target string) is kept and the rest discarded. The best copy is copied
several times again, with mutations, the best of these copies is kept,
and the process is repeated until the target string is reached (or
your patience expires). The closest biological system to this is
sequential selection of bacteria by exposing populations to sequential
concentrations/types of antibiotics and propagating a _single_
bacteria between exposures, or selection for catalytic RNA by making
multiple mutant copies of a random (or known, but non catalytic)
sequence, selecting a _single_ highly performing sequence, then
repeating the round of selection and mutation until you have a highly
active sequence. These are all good examples of selection and its
efficay, but not exemplars of natural selection in the wild.
To re-iterate, Dawkins program shows how selection is far more
efficacious than random search, and there are real world biological
analogs of his computer search. However, it is NOT a _simulation_ of
natural selection (nor was it intended to be). Importantly, Dawkins
program CANNOT show Haldane's dilemma, as the rate of fixation of
beneficial mutations is _always_ 1 generation, regardless of whether
there are one, two or more beneficial mutations present in a single
organism (recall that Haldane's dilemma says it will take longer to
fix two simultaneous beneficial mutations than to fix one beneficial
mutation).
It is important to remember that in Dawkins program that the breeding
population is ALLWAYS _one_. What ReMine is changing is the number of
offspring that single breeding "organism" has. Naturally, when you
have only one breeding organism, and you reduced the number of
offspring to 6 or so, and then breed from only one of those offspring,
the rate of change is going to be very, very slow, as it takes a long
time for beneficial mutations to turn up. However, when such a
mutation (or a pair of mutations, or 3 mutations) turn up in an
offspring, they will be fixed in _one_ generation.
The fact that Mr. ReMine thinks that his manipulations of Dawkins
program says anything at all about Haldane's dilemma is measure of his
profound misunderstanding of Haldane's dilemma, and evolution in
general.
Cheers! Ian
=====================================================
Ian Musgrave Peta O'Donohue,Jack Francis and Michael James Musgrave
reyn...@werple.mira.net.au http://werple.mira.net.au/~reynella/
Southern Sky Watch http://www.abc.net.au/science/space/default.htm
Steven J.
> I responded to it at the time, and my posts from that era can still be
> recovered from Google.com "Advanced Groups Search" by searching for the
> author:
> "laser...@my-deja.com".
>
> For example, below is a copy (from 1999) of my first response to his
> misrepresentations.
>
>
> Subject: Dawkins Simulation & Haldanes Dilemma
> Newsgroups: talk.origins
> Date: 1999/09/28
>
>
> Dawkins Simulation & Haldane's Dilemma
>
> Before I get to the technical details I must give some background.
>
-- [snip]
>
> Ian Musgrave recently posted the following point. Due to the peculiar
> inner construction of the Dawkins computer simulation, the reproduction
> rate and population size are directly linked, unlike in nature. Thus,
> when one lowers the reproduction rate by a factor of sixteen, that
> lowers the population size by a factor of sixteen. Therefore he suggests
> I had done something unrealistic to this simulation. Just who is he
> kidding! Dawkins's simulation is unrealistic from the get-go -- wildly
> in favor of evolution. No mere factor of sixteen changed that.
>
unrealistic. It assumes, in its most obviously inaccurate aspect,
that the selection process has some preset, invariable goal. The
branching "tree of life" of evolution would not be possible if that
were so. I also assumes unrealistically high selection pressures.
And it has other unrealistic aspects.
However, the weasel applet is designed, not to model evolution _in
toto_, but to show the difference between cumulative selection (in
which one is allowed to keep and build on past partial successes) and
"one shot" selection, in which one must arrive at a complete solution
in a single try. It shows the disparity in the number of tries needed
under these different models of selection, and that is its only
purpose.
>
> Yet Musgrave's post completely omitted that. He selectively quoted from
> my book and SELECTIVELY OMITTED the vast bulk of its material revealing
> the many unrealistic pro-evolution features of this simulation. There is
> an abundance of that material, so abundant that much of it is neatly
> stacked in densely packed, bulleted lists. It is impossible to read
> that section of my book (six pages) and not be overwhelmed by its
> presence. That material is directly relevant to the issue he raised --
> Is the simulation unfair to evolution? My book is on top of his point.
> It is hammering all over the point. Yet he omitted any mention of it!
> He omitted any vague reference to it. He recklessly misrepresented my
> book.
>
Whether Dawkins's model was an accurate model of how cumulative
selection worked in nature (again, a claim he did not make) is
irrelevant to the question of whether your version of the model is an
accurate model.
>
> And once again, not one evolutionist objected. Not one. Not one said
> the obvious, "Mr. Musgrave, I have seen ReMine's material, and you have
> unfairly omitted the guts of his argument. Wouldn't it be better to
> deal with his material as it actually is, rather than manufacture some
> neutered version of it? Wouldn't it have been better to at least
> mention ReMine's points that have a direct bearing on your issue?" No,
> not one evolutionist objected to Musgrave's misrepresentation of my
> work. On the contrary, Tim Ikeda (in a followup post) even found it
> humorous.
>
Are you denying that you implied that you model showed that mutation
and natural selection would not work? Are any defects in Dawkins's
model any justification whatsoever for defects in yours?
>
> I hope there will someday be an honest debate about origins (and my
> material), but that is impossible in an environment that fosters and
> embraces rampant misrepresentation. That is why thoughtful folks
> routinely avoid such places, as I do.
>
And all this time I thought it was because you kept getting stomped
into the ground by better-informed posters.
Bobby D. Bryant
wrote:
> Dawkins program was not designed to be a simulation of natural
> selection. Dawkins says this explicitly and mentions some of the
> reasons that it isn't, and what a real simulation would entail.
>
> What it _is_ is a demonstration of the efficacy of selection over
> random search (which it does quite admirably).
I've posted this before, but new lurkers may want to check out
http://www.cs.utexas.edu/users/bdbryant/talk-origins/ga-on-tsp/index.html
The page has some screenshots of a GA that I hacked to show
side-by-side shots of the progress of selected search vs. random
search on the Travelling Salesman Problem.
It's not a very good model of biology either, but it has a nice
graphical output, and unlike WEASEL it does not rely on a
pre-specified answer. (The "fitness" rule is simlply shorter path =
better.)
Also, though it may not be possible to map Haldane's Delimma onto
what this solver does, I _have_ run it with lots of different
population sizes, and the general rule is that larger populations
find better solutions.
Bobby Bryant
Austin, Texas
Roy Thearle
in news:<Xns92667BF88DFC4...@206.191.138.115> you wrote
> Let me use Musgrave's words to make the point:
>
> "Trying to compare the substitution rate
> in a population of [N=100] individuals
> with the substitution rate in a population
> of between 10,000 to 100,000 individuals
> is a pretty big blunder to make"
> (from Musgrave's post, brackets mine,
> N=100 from Dawkins's simulation)
But those aren't Musgrave's words. Musgrave writes:
"Trying to compare the appearance rate of beneficial mutations in a
population of 5 individuals with the substitution rate of beneficial
mutations in a population of between 10,000 to 100,000 individuals
is a pretty big blunder to make"
Musgrave's main point was that you had changed the population size in
the simulation from the default of 100 down to 5, possibly through
ignorance of
how the program worked, and then commented on the
slow rate of substitution.
Yet in your quote above, you appear to have changed Musgrave's text to
hide
this point, and make it appear that he is referring to the default
setting in the simulation.
Do you have an explanation?
Roy
John Wilkins
Could someone do a random target generator for a mustelidian algorithm
that changed the targets over time within a certain locale, and see how
it tracks them?
One of the "objections" to WEASEL is that the target is prespecified (so
what, I wonder?), so why not have them a little more realistic, with
gradual change to the targets from a random starting point?
--
John Wilkins
Sweet Analytics, 'tis thou hast ravished me [Marlowe's Faust]
Ian Musgrave & Peta O'Donohue
Address altered to avoid spam, delete RemoveInsert
On Sun, 11 Aug 2002 09:23:26 +0000 (UTC), rthe...@hotmail.com (Roy
Thearle) wrote:
>Mr (Dr?) Remine,
>in news:<Xns92667BF88DFC4...@206.191.138.115> you wrote
>
>> Let me use Musgrave's words to make the point:
>>
>> "Trying to compare the substitution rate
>> in a population of [N=100] individuals
>> with the substitution rate in a population
>> of between 10,000 to 100,000 individuals
>> is a pretty big blunder to make"
>> (from Musgrave's post, brackets mine,
>> N=100 from Dawkins's simulation)
>
>But those aren't Musgrave's words. Musgrave writes:
>
>"Trying to compare the appearance rate of beneficial mutations in a
> population of 5 individuals with the substitution rate of beneficial
> mutations in a population of between 10,000 to 100,000 individuals
> is a pretty big blunder to make"
Note the unmarked change of "appearance rates" to "substitution
rates". This is not a trivial change. In the program, the substitution
rate is _always_ one. That is, any beneficial mutation or group of
beneficial mutations is fixed automatically in one generation. Thus
this program CANNOT demonstrate Haldane's dilemma.
>Musgrave's main point was that you had changed the population size in
>the simulation from the default of 100 down to 5, possibly through
>ignorance of how the program worked, and then commented on the
>slow rate of substitution.
It's worse than that 100, 5 etc is not the "population size", but the
"offspring number". In this program there is _one_ and one only,
breeding "organism" in the population, it has five offspring (or 100),
only _one_ of these offspring is selected to be the new breeder, and
it has five offspring (or 100), then only one of _these_ offspring is
chosen to be the new breeder, and so on.
As you can see, in a population where there is only one breeding
organism, and only 5 or so offspring, evolution will be slow because
it takes a blasted long time for a favorable mutation to arise in such
a small population. When population biologist talk about "small"
populations they are generally talking about 100-200 _breeding_
individuals. As you can see, ReMine's manipulations of Dawkins program
are irrelevant to either Haldane's dilemma or evolution.
>Yet in your quote above, you appear to have changed Musgrave's text to
>hide this point, and make it appear that he is referring to the default
>setting in the simulation.
>
>Do you have an explanation?
A _very_ good question
Cheers! ian
Bobby D. Bryant
> Mr (Dr?) Remine,
...
> Yet in your quote above, you appear to have changed Musgrave's text to
> hide this point, and make it appear that he is referring to the default
> setting in the simulation.
>
> Do you have an explanation?
Yeah:
"I am a creationist and this is the norm for my trade."
Bobby Bryant
Austin, Texas
p.s. - Looks like you got a couple of words swapped 'round in your
subject line, Mr. ReMine, so I fixed it for you.
Andy Groves
I would be particularly interested to hear Walter's explanation. It's
rather excessive - even for a creationist - to re-enter a usenet forum
after a break, and deliberately falsify someone's words in your very
first post.
Andy
John Wilkins
You think? How often do they unmark snips to do that very same thing? I
think that is a special case of the same phenomenon. Or is it the
no-wait immediate service after a long absence aspect that is unusual
here?
Roy Thearle
> On Sun, 11 Aug 2002 09:23:26 +0000 (UTC), rthe...@hotmail.com (Roy
> Thearle) wrote:
>
> >Mr (Dr?) Remine,
> >in news:<Xns92667BF88DFC4...@206.191.138.115> you wrote
> >
> >> Let me use Musgrave's words to make the point:
> >>
> >> "Trying to compare the substitution rate
> >> in a population of [N=100] individuals
> >> with the substitution rate in a population
> >> of between 10,000 to 100,000 individuals
> >> is a pretty big blunder to make"
> >> (from Musgrave's post, brackets mine,
> >> N=100 from Dawkins's simulation)
> >
> >But those aren't Musgrave's words. Musgrave writes:
> >
> >"Trying to compare the appearance rate of beneficial mutations in a
> > population of 5 individuals with the substitution rate of beneficial
> > mutations in a population of between 10,000 to 100,000 individuals
> > is a pretty big blunder to make"
>
> Note the unmarked change of "appearance rates" to "substitution
> rates". This is not a trivial change. In the program, the substitution
> rate is _always_ one. That is, any beneficial mutation or group of
> beneficial mutations is fixed automatically in one generation. Thus
> this program CANNOT demonstrate Haldane's dilemma.
Yup, I n oticed that and I agree that the original Dawkins' algorithm
cannot
show Haldane's dilemma. However, since understanding that aspect of
the
changed quote requires some knowledge of genetics, I decided to
concentrate on
the blatant, obvious-to-anyone-who-can-count, change.
> >Musgrave's main point was that you had changed the population size in
> >the simulation from the default of 100 down to 5, possibly through
> >ignorance of how the program worked, and then commented on the
> >slow rate of substitution.
>
> It's worse than that 100, 5 etc is not the "population size", but the
> "offspring number". In this program there is _one_ and one only,
> breeding "organism" in the population, it has five offspring (or 100),
> only _one_ of these offspring is selected to be the new breeder, and
> it has five offspring (or 100), then only one of _these_ offspring is
> chosen to be the new breeder, and so on.
Um, yes, I know. I've written programs like this (though mine didn't
have this drawback).
Roy
Mitchell Coffey
The unusual aspect here is that ReMine, as I recall, holds an
academic-style appointment in an organization, the Discovery
Institute, that pretends to scientific standards.
In real science a swift correction and explaination would be in order.
Now, in the United States these days there's been a rise of
pseudo-academic advocacy "institutes" that ignore academic standards
as appropriate to their purpose. I'm thinking of the American
Enterprise Institute and the Heritage Foundation specifically. If
Discovery holds itself to the standards of The Bell Curve, etc., it
should be noted.
Mitchell Coffey
mREMOVE....@CAPSstarpower.PLEASEnet
Mitchell Coffey
James Acker
[deletion]
: For example, below is a copy (from 1999) of my first response to his
: misrepresentations.
: -- Walter ReMine
: Fellow with Discovery Institute
: Center for the Renewal of Science and Culture
: _The Biotic Message_
: http://www1.minn.net/~science
While I will note that another t.o. poster actually called
the CRSC and confirmed that Mr. ReMine has received funding from the
CRSC, it is curious that they have not ever update their Fellows
page to include him.
http://www.discovery.org/crsc/fellows/index.html
The CRSC regularly updates their home page,
http://www.discovery.org/crsc/
(most recently with an article by Wesley Smith dated July 16, 2002),
so why the omission of Fellow ReMine? One possibility would be that
there is a difference between someone that has received a research
fellowship from the CRSC and someone who has actually been named a
Fellow of the CRSC. If so, then Mr. ReMine is misrepresenting
himself in the signature file above.
If, on the other hand, the CRSC has merely been lax in
updating their Fellows page, I would be glad to note this oversight
to them. Mr. ReMine should be certainly be recognized by the CRSC
in this manner if this is an honor accorded to all other CRSC Fellows.
1999 posting excerpt follows:
-------
: From: laser...@my-deja.com (laser...@my-deja.com)
: Subject: Dawkins Simulation & Haldanes Dilemma
: Newsgroups: talk.origins
: Date: 1999/09/28
: Move forward again to a month ago. James Acker began posting
: misrepresentations of my material, and I posted my objection. Acker then
: publicly vowed to CONTINUE his [mis-]presentations even though he hasn't
: read my book.
Update: I have read the book.
: The remarkable thing, the important point, is that through-out all this
: NOT ONE EVOLUTIONIST OBJECTED. Not one! Not one evolutionist said the
: obvious, "Messrs. -- Elsberry, Dorman, Williams, and Acker -- you really
: ought to give an honest reading to the material you are criticizing, at
: a bare minimum, wouldn't that be the scholarly and honest thing to do?"
Having read the book, there were no credible objections that
could be made.
[remainder deleted]
On the other hand, I will again note that Mr. ReMine never
read any of the peer-reviewed references regarding the fossil record
of cetacean ancestry which were repeatedly described to him and
easily found in mainstream scientific journals.
Jim Acker
*** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** ***
Jim Acker
jac...@gl.umbc.edu
"Since we are assured that an all-wise Creator has observed the
most exact proportions, of number, weight, and measure, in the
make of all things, the most likely way therefore, to get any
insight into the nature of those parts of the creation, which
come within our observation, must in all reason be to number,
weigh, and measure." - Stephen Hales
Walter ReMine
> Musgrave suggests it is 'unrealistic' to use a small
> population size in the simulation. Doesn't that
> strike you as odd, since any other day of the year
> evolutionists promote Dawkins's simulation to an
> unsuspecting public. Let me use Musgrave's words
> to make the point:
>
> "Trying to compare the substitution rate
> in a population of [N=100] individuals
> with the substitution rate in a population
> of between 10,000 to 100,000 individuals
> is a pretty big blunder to make"
> (from Musgrave's post, brackets mine,
> N=100 from Dawkins's simulation)
>
> If Musgrave believes his own argument, then Dawkins
> made a "pretty big blunder". If he wants to make
> charges, they must be placed on Dawkins first.
> Followed by evolutionists as a whole, because they
> fully promoted this simulation and failed to inform
> the public about its many reckless pro-evolution
> features.
Musgrave's recent response is:
> Yes, but then they are NOT my words when you do that,
> (check to original for the real usage)
Followed by Roy Thele's charge:
> But those aren't Musgrave's words.
Musgrave doesn't clear up the charges, but rather continues it with a
misleading suggestion that *I* changed his words in a non-trivial way:
> Note the unmarked change of "appearance rates"
> to "substitution rates". This is not a trivial change.
[Note: For the record, my post (from 1999, quoted above) did not misquote
Musgrave -- those are his actual words in his 1999 post. In his recent
webpage, he has since CHANGED his wording slightly, so don't blame me for
not foretelling his future back in 1999.]
I did not misquote Musgrave -- the fact that the brackets are mine is
clearly marked. I merely plugged-in Dawkins's figure (N=100) to put
Musgrave's hyperbolic comment into perspective -- his sentence is
substantially unaffected, suggesting that Dawkins made a "pretty big
blunder". Dawkins (in _The Blind Watchmaker_) bragged about the speed of
his WEASEL program, he gave specific figures for it. (In three separate
trials the target phrase is reached in 43, 64, and 41 generations.) The
public is drawn in by Dawkins's illusion, and he did little to disabuse
them of its power. (The many facets of that illusion are disassembled in
my book.) The point at the moment is that Dawkins's results became the de
facto standard of comparison for the public.
Indeed, Dawkins's SMALL population size MAGNIFIES THE ILLUSION -- for if
evolution can proceed this fast (for a single gene, such as hemoglobin,
which was the context of Dawkins's discussion), then imagine how
EXCEEDINGLY MUCH FASTER the evolution would be in a LARGE population and
with MANY genes evolving at this rate! Or so the public was led to
believe. Dawkins's program was a whopper illusion-maker, and other
evolutionists likewise did little to disabuse the public of it.
> Dawkins didn't make any blunder, as Dawkins was never
> making any claim about Haldane's dilemma.
Dawkins didn't mention Haldane's Dilemma, because evolutionists obscured it
into oblivion and prematurely brushed it aside -- ESPECIALLY when
addressing the generally public. If Dawkins was even aware of Haldane's
Dilemma, he had nothing to gain from alerting the public to this sleeping
dragon. The Haldane-based limitation of approximately 1667 beneficial
nucleotides for human evolution was not something evolutionists wanted
taught publicly -- and indeed they never did (until after my book). Yet
his WEASEL program furthered the illusion that adaptive evolution could be
exceedingly fast (and unbound by any Haldane speed limit). Dawkins's
"blunder" (Musgrave's word not mine) is that he created an illusion.
My dismantling of Dawkins's illusion is entirely fair-play. To the public,
his WEASEL program is easily the most widely known computer model of
natural selection. Yet it contains many, many unrealistic factors ALL
WILDLY IN FAVOR OF EVOLUTION!
One major argument between Musgrave and me goes as follows. Musgrave says
(correctly) that "population size matters", and he argues (explicitly or
implicitly, depending on the post) that a small population size (as with
N=6) is unrealistic. I've always agreed, it's wildly unrealistic, and
there are over a dozen additional factors about Dawkins's WEASEL program
that are wildly unrealistic -- but they are ALL UNREALISTIC IN FAVOR OF
EVOLUTION AND ITS SPEED!!! Yes, even N=6 FAVORS THE SPEED OF EVOLUTION, as
follows:
[Begin quote from my post of Sep 28, 1999]
[END quote from my post of Sep 28, 1999]
Evolutionists still haven't identified anything about my use of Dawkins's
WEASEL program that DIS-FAVORS evolution, much less tips the scales
overall. It's all wildly in favor of evolution. Therefore, it is eye-
opening to see it produces a speed comparable to the Haldane speed limit.
Bobby D. Bryant
Yes, Ian did say "substitution rate" in his 1999 post. But you've
confused everyone by quoting the 1999 post in reply to a post where
Ian gave a link to his _revised_ version of the argument, which does
_not_ say "substitution rate". You need to mark such context
switches clearly.
However, Ian's words quoted above were a reply to _you_, not to Roy.
IMO he erred by not pointing out the problem in his separate reply
to Roy, but your quote mining above raises questions about whether
you're trying to address the facts or misrepresent them. Ian's
words are exactly correct in the context where he used them.
And that context addressses the bottom line of the matter: your
substitution of "[N=100]" for Ian's "5" (both in his original and in
his revision) is not a legitimate 'quote' of Ian, whether you use
brackets and mention the substitution parenthetically or not. If
you want to quote Ian, don't substitute in something derived from a
program's default settings. Honest people do use bracketized
comments to clarify quotes sometimes, but using them for unmotivated
changes to an author's words is inexcusable. If you want to discuss
what Ian said, quote what Ian said.
And if you want to discuss what he said three years ago instead of
what he said yesterday, make it clear that you're switching
contexts, and provide a google link so everyone can see what you're
talking about.
> I did not misquote Musgrave -- the fact that the brackets are mine
> is clearly marked. I merely plugged-in Dawkins's figure (N=100)
> to put Musgrave's hyperbolic comment into perspective -- his
> sentence is substantially unaffected, suggesting that Dawkins made
> a "pretty big blunder".
Fogive me, but I view this as outright dishonesty. If you want to
put someone's comments into perspective, do it in your own words
rather than modifying what that person actually said.
> Dawkins (in _The Blind Watchmaker_) bragged about the speed of his
> WEASEL program, he gave specific figures for it. (In three
> separate trials the target phrase is reached in 43, 64, and 41
> generations.) The public is drawn in by Dawkins's illusion, and
> he did little to disabuse them of its power. (The many facets of
> that illusion are disassembled in my book.) The point at the
> moment is that Dawkins's results became the de facto standard of
> comparison for the public.
And here you indicate that you have _completely_ _missed_ Ian's
point, both in 1999 and 2002.
> Indeed, Dawkins's SMALL population size MAGNIFIES THE ILLUSION --
> for if evolution can proceed this fast (for a single gene, such as
> hemoglobin, which was the context of Dawkins's discussion), then
> imagine how EXCEEDINGLY MUCH FASTER the evolution would be in a
> LARGE population and with MANY genes evolving at this rate! Or so
> the public was led to believe. Dawkins's program was a whopper
> illusion-maker, and other evolutionists likewise did little to
> disabuse the public of it.
I don't know whether anyone "led" the public to believe that or not,
but FWIW in my own work with genetic algorithms I find that larger
population sizes almost always help find a better solution within a
given number of generations. On that basis I would conjecture that
what you identify as a misleading of the public is in fact correct.
>> Dawkins didn't make any blunder, as Dawkins was never making any
>> claim about Haldane's dilemma.
>
> Dawkins didn't mention Haldane's Dilemma, because evolutionists
> obscured it into oblivion and prematurely brushed it aside --
> ESPECIALLY when addressing the generally public. If Dawkins was
> even aware of Haldane's Dilemma, he had nothing to gain from
> alerting the public to this sleeping dragon. The Haldane-based
> limitation of approximately 1667 beneficial nucleotides for human
> evolution was not something evolutionists wanted taught publicly
> -- and indeed they never did (until after my book). Yet his
> WEASEL program furthered the illusion that adaptive evolution
> could be exceedingly fast (and unbound by any Haldane speed
> limit). Dawkins's "blunder" (Musgrave's word not mine) is that he
> created an illusion.
No, you're the one who's creating the illusion here, by trying to
interpret Dawkins' program as a model for something Dawkins didn't
offer it for.
> My dismantling of Dawkins's illusion is entirely fair-play. To
> the public, his WEASEL program is easily the most widely known
> computer model of natural selection. Yet it contains many, many
> unrealistic factors ALL WILDLY IN FAVOR OF EVOLUTION!
IMO it works well as a simple example of how a selection factor can
make random mutations deliver results _much_ faster than independent
random guesses (that ever-popular strawman that stands in for
evolution in most creationist arguments).
I personally would have rather seen him use a more conventional
genetic algorithm rather than the simple WEASEL program, but he had
a specific point to make and apparently decided to optimize for
simplicity rather than biological accuracy. I do the same thing
sometimes: I don't view most genetic algorithms as very good models
for nature at all, but I _do_ think they demonstrate the principle
that random mutations, filtered by selection and fed back into the
the next generation via a feedback loop, can have the sort of
effect that the theory of evolution requires. But that does not
mean that I think it's legitimate to conclude anything you please
from observing a genetic algorithm's behavior.
Perhaps before we go any further you should clarify your views on a
couple of things:
a) What point was Dawkins trying to make when he described WEASEL?
b) Is WEASEL a good way to make that point?
It looks to me like you're trying to do for Dawkins what you've done
for Ian. Ian's words are only valid for Ian's point. Dawkins' model
is only good for Dawkins' point. If you want to make your own
point, present your own words and model rather than trying to twist
other people's around.
> One major argument between Musgrave and me goes as follows.
> Musgrave says (correctly) that "population size matters", and he
> argues (explicitly or implicitly, depending on the post) that a
> small population size (as with N=6) is unrealistic. I've always
> agreed, it's wildly unrealistic, and there are over a dozen
> additional factors about Dawkins's WEASEL program that are wildly
> unrealistic -- but they are ALL UNREALISTIC IN FAVOR OF EVOLUTION
> AND ITS SPEED!!!
But the actual question is, what does WEASEL say about the
_relative_ speed of mutations with selection vs. mutations without
selection? Don't try to read more in to it than what the author
actually suggests. (Did you answer my two questions above yet?)
> Yes, even N=6 FAVORS THE SPEED OF EVOLUTION, as follows:
>
> [Begin quote from my post of Sep 28, 1999]
>
> The population is tiny -- its cost of substitution is also tiny,
> at 5. And that tiny cost FAVORS the speed of evolution. It is
> one-sixth the cost that Haldane estimated as the average for
> evolution -- 30.
Enough of this obfuscation! Was Dawkins trying to address Hadane's
dilemma or not?
> This simulation incurs a tiny cost, and for a
> given reproduction rate that makes for faster evolution. In
> Dawkins's simulation, the small population size favors evolution.
> (And I have not even begun to recount here the well over a dozen
> other ways this simulation favors evolution.) Nonetheless, with
> all these advantages, this simulation is less than five times
> faster than the rate known as Haldane's Dilemma.
>
> Now my opponents ought consider the following question: For what
> reason would a small population DIS-favor the evolutionary
> simulation? If you try to find such a reason, you will fail.
In general smaller populations can harbor less diversity, and
diversity is absolutely essential for evolution. That's why we
cosistently see that in genetic algorithms applied to non-trivial
problems, bigger populations are advantageous.
> You
> will find the usual reasons are neatly disabled by the design of
> Dawkins's simulation. His simulation disallows the usual
> disadvantages of a small population. I claim that this small
> population (with N=6) in fact has the advantages of a large
> population -- without the high substitution cost of a large
> population. Put simply, Dawkins's simulation is peculiarly
> designed to have the advantages of small and large populations *at
> the same time*. It has the best of both! A natural population
> cannot possibly do that, but the simulation is unrealistically
> designed to do that. And it is all in FAVOR of evolution.
Have you answered my two questions above yet?
If you want a model that addresses Haldane's Dilemma, offer us a
model yourself rather than criticizing a model that wasn't intended
to address it.
> Take it as a challenge. Think of some disadvantage of a small
> population, then carry it through to see how Dawkins's design gets
> around that disadvantage. After you've had this "Aha!"
> experience, then notice that the reasons are clearly called out in
> my section on this simulation, and explained in my book, and that
> Musgrave unscrupulously omitted any mention of them.
>
> [END quote from my post of Sep 28, 1999]
>
> Evolutionists still haven't identified anything about my use of
> Dawkins's WEASEL program that DIS-FAVORS evolution, much less tips
> the scales overall. It's all wildly in favor of evolution.
> Therefore, it is eye- opening to see it produces a speed
> comparable to the Haldane speed limit.
You raise the question that always comes up when dealing with
creationists: "Is this guy clueless, dishonest, or both?"
You have completely tuned out what Ian has been pointing out for
three years, and you're trying to support your position with a
conspiracy theory rather than by addressing the issues. You claim
that WEASEL doesn't address Haldane's Dilemma and Ian claims that
it _can't_ address Haldane's Dilemma. So what's the problem?
Bobby Bryant
Austin, Texas
Dirk Murcray
> Dawkins didn't mention Haldane's Dilemma, because evolutionists obscured it
> into oblivion and prematurely brushed it aside -- ESPECIALLY when
> addressing the generally public. If Dawkins was even aware of Haldane's
> Dilemma, he had nothing to gain from alerting the public to this sleeping
> dragon. The Haldane-based limitation of approximately 1667 beneficial
> nucleotides for human evolution was not something evolutionists wanted
> taught publicly -- and indeed they never did (until after my book).
Yet oddly, Haldane's Dillema was taught along with standard
Hardy-Weinberg equations and Punnet Square excercises in my college
biology and physical anthropology classes in the mid-80's, several
years before The Biotic Message was published. Since the pace of
evolution in real life obviously does not square with Haldane's
numbers on paper, it was simply recognized (as did Haldane) that
population geneticists still had much to learn regarding evolutionary
processes.
Bobby D. Bryant
Yet another example of my recent soapbox topic: when people notice
something that doesn't jibe with expectations, creationists cry
"goddidit!" and scientists start looking for explanatory mechanisms.
BTW, I especially like ReMine's claim that Dawkins was helping
suppress a fact that he might not even have been aware of. (What is
it with pseudoscientists and their paranoid conspiracy theories,
anyway?)
Bobby Bryant
Austin, Texas
Roy Thearle
> I did not misquote Musgrave -- the fact that the brackets are mine is
> clearly marked. I merely plugged-in Dawkins's figure (N=100) to put
> Musgrave's hyperbolic comment into perspective -- his sentence is
> substantially unaffected, suggesting that Dawkins made a "pretty big
> blunder".
You did misquote Musgrave.
Musgrave was clearly referring to your alteration of the parameter of
the program; by re-inserting the default value you are obscuring that fact.
That you used [] while doing so is no excuse.
Roy
Roy Thearle
> Dawkins admits explicitly that his simulation is in many respects
> unrealistic.
[rest excised]
On a related note, I've been trying to remember what Dawkins
originally wrote in "The Blind Watchmaker", as I won't have access to
my copy for a while.
Is it correct that the original program didn't have a population at
all, but just had a single string for which every letter that didn't
match the target was mutated in every generation?
If so, this makes all references to populations and Haldane's dilemma
somewhat superfluous.
Roy
Adam Marczyk
news:Xns92692A9E838CE...@206.191.138.115...
[snip]
> Dawkins (in _The Blind Watchmaker_) bragged about the speed of
> his WEASEL program, he gave specific figures for it. (In three separate
> trials the target phrase is reached in 43, 64, and 41 generations.)
Yes. He did. That was the point of the program - to illustrate the difference
between cumulative selection and random single-step selection, to dismantle the
creationist argument that "evolution is as likely as a tornado blowing through a
junkyard and producing a 747" - an argument repeated by creationists as recent
as William Dembski. Dawkins' point is that evolution is a process of small
changes accumulating, not a one-step random combination of parts. That is *all*
the weasel program is designed to demonstrate - the difference between
cumulative and single-step selection - and Dawkins never stated or implied that
he intended it to demonstrate anything else. Naturally, that has not stopped
creationists such as yourself from aggressively misunderstanding it.
--
And I want to conquer the world,
give all the idiots a brand new religion,
put an end to poverty, uncleanliness and toil,
promote equality in all of my decisions...
--Bad Religion, "I Want to Conquer the World"
http://www.ebonmusings.org ICQ: 8777843 PGP Key ID: 0x5C66F737
Ian Musgrave & Peta O'Donohue
Address altered to avoid spam, delete RemoveInsert
And they are not.
>Followed by Roy Thele's charge:
>> But those aren't Musgrave's words.
>
>Musgrave doesn't clear up the charges, but rather continues it with a
>misleading suggestion that *I* changed his words in a non-trivial way:
>> Note the unmarked change of "appearance rates"
>> to "substitution rates". This is not a trivial change.
>
>[Note: For the record, my post (from 1999, quoted above) did not misquote
>Musgrave -- those are his actual words in his 1999 post. In his recent
>webpage, he has since CHANGED his wording slightly, so don't blame me for
>not foretelling his future back in 1999.]
This was in response to a post about my Web page, some people might
find it a strange fancy that one should quote from the website when
replying to a post from the website. Especially when the post
explicitly says the web site supersedes earlier posts.
>I did not misquote Musgrave -- the fact that the brackets are mine is
>clearly marked. I merely plugged-in Dawkins's figure (N=100) to put
>Musgrave's hyperbolic comment into perspective -- his sentence is
>substantially unaffected, suggesting that Dawkins made a "pretty big
>blunder".
And thereby changing the meaning of the sentence.
[snip]
>> Dawkins didn't make any blunder, as Dawkins was never
>> making any claim about Haldane's dilemma.
>
>Dawkins didn't mention Haldane's Dilemma, because evolutionists obscured it
>into oblivion and prematurely brushed it aside
You are wrong, but that is not the point. YOU have claimed,
repeatedly, that Dawkins program demonstrates Haldane's dilemma.
Haldane's Dilemma is NOT "evolution goes slow", it is that two or more
simultaneous beneficial mutations will take as long to fix in a
population as two or more sequential beneficial mutations. In Dawkins
program beneficial mutations, no matter how many, are _always_ fixed
in one generation. It cannot demonstrate Haldane's Dilemma.
[snip]
>One major argument between Musgrave and me goes as follows. Musgrave says
>(correctly) that "population size matters", and he argues (explicitly or
>implicitly, depending on the post) that a small population size (as with
>N=6) is unrealistic. I've always agreed, it's wildly unrealistic, and
>there are over a dozen additional factors about Dawkins's WEASEL program
>that are wildly unrealistic -- but they are ALL UNREALISTIC IN FAVOR OF
>EVOLUTION AND ITS SPEED!!! Yes, even N=6 FAVORS THE SPEED OF EVOLUTION, as
>follows:
There is only one breeding string, regardless of N, N is the number of
offspring that the breeding string has. N=6 will slow the rate of
evolution in Dawkins program because it limits the rate at which
mutations appear.
>[Begin quote from my post of Sep 28, 1999]
>
>The population is tiny -- its cost of substitution is also
>tiny, at 5.
All this shows is that you don't understand Haldane's Dilemma, or what
the program does. Mutations are ALLWAYS fixed in one generation,
regardless of the number of offspring. Cost of substitution is
irrelevant to fixation rate (which is always constant at 1) in Dawkins
program.
Yes, by having the single breeding string have only 6 offspring, you
can slow the rate of evolution in Dawkins simulations right down,
because you affect the mutation supply. That you think this has
_anything_ to do with Haldane's Dilemma shows your profound
misunderstanding of both Haldane's dilemma and Dawkins program.
[snip profound misunderstanding]
Dirk Murcray
In order for the uniformed public to accept their baseless theories,
they first need to discredit mainstream science as elitist, dogmatic,
and even worse, bent on purging society of religion. Based on the
declining emphasis on quality science education in recent years, I
fear they are succeeding.
Bobby D. Bryant
wrote:
> You are wrong, but that is not the point.
Still, it's interesting to see how much effort he spends nitpicking
minor faux pas in what you say (or don't say), how much effort he
spends doctoring up quotations and contexts, how many accusations of
conspiracy he makes, and how much flak he fires at the biological
realism of the WEASEL program -- all without ever even _attempting_
to address the basic point you made three years ago.
IMO even if special creation really happened the creationists
wouldn't be able to make a case for it, because they're all so
frikkin clueless about how to construct and support a position on
anything.
Bobby Bryant
Austin, Texas
Noelie S. Alito
news:68494300.02081...@posting.google.com...
Well, the-person-who-posts-as-ReMine has very frequently declared
that talk.origins is "a culture of misrepresentation". Maybe he's just
trying to fit in to the culture as he sees it?
[Awaiting the "whoosh" as the point zooms past ReMine's head.]
Noelie
--
"Rhyming with 'goalie' for over 42 years."
Bobby D. Bryant
> "Roy Thearle" <rthe...@hotmail.com> wrote in message
> news:68494300.02081...@posting.google.com...
>> Musgrave was clearly referring to your alteration of the parameter of
>> the program; by re-inserting the default value you are obscuring that fact.
>>
>> That you used [] while doing so is no excuse.
>
> Well, the-person-who-posts-as-ReMine has very frequently declared
> that talk.origins is "a culture of misrepresentation". Maybe he's just
> trying to fit in to the culture as he sees it?
>
> [Awaiting the "whoosh" as the point zooms past ReMine's head.]
I don't think there's enough airspace for all the points trying to
whoosh him right now. Would it be OK with you if we schedule yours
to whoosh about a quarter mile to the east? I know that's not very
good, but you'll understand we have to arrange things on a first
come - first served basis.
> Noelie
> --
> "Rhyming with 'goalie' for over 42 years."
Would you mind sharing some of the better rhymes you two have come
up with?
Bobby Bryant
Austin, Texas
--
Not quite rhyming with "wiseacre" for as long as I can remember.
James Acker
: Walter ReMine <sci...@nospam.net> wrote:
: [deletion]
: http://www.discovery.org/crsc/fellows/index.html
: http://www.discovery.org/crsc/
I have communicated with Jay Richards of the renamed Center
for Science and Culture, who has indicated that an individual who
receives a research fellowship from the CSC is indeed a Fellow of
the CSC, as Mr. ReMine notes in his signature file. Mr. Richards
indicated that their Fellows list is not complete, which I think is
an oversight on their part; they should at least indicate that the
list is incomplete.
Mr. Richards also indicated that Mr. ReMine had received
his fellowship for research on Haldane's Dilemma. Hopefully this
research funding will result in several paper submissions to peer-reviewed
scientific journals.
It would be fun to inform Mr. Richards about the current
assessment of Mr. ReMine's understanding of Haldane's Dilemma discussed
in this thread, but I won't do that.
[remainder deleted]
Derek Stevenson
news:ajdtnq$upu$1...@news.umbc.edu...
> I have communicated with Jay Richards of the renamed Center
> for Science and Culture, who has indicated that an individual who
> receives a research fellowship from the CSC is indeed a Fellow of
> the CSC, as Mr. ReMine notes in his signature file. Mr. Richards
> indicated that their Fellows list is not complete, which I think is
> an oversight on their part; they should at least indicate that the
> list is incomplete.
>
> Mr. Richards also indicated that Mr. ReMine had received
> his fellowship for research on Haldane's Dilemma. Hopefully this
> research funding will result in several paper submissions to
peer-reviewed
> scientific journals.
As a dear, departed friend of mine used to observe, "Sarcasm is the lowest
form of wit, and belongs in the gutter with those who use it."
No, it never stopped *me*, either.
> It would be fun to inform Mr. Richards about the current
> assessment of Mr. ReMine's understanding of Haldane's Dilemma discussed
> in this thread, but I won't do that.
Why ever not? Surely such an assessment would only strengthen the
Institute's perception of his suitability to undertake such a study.
Scott
> into oblivion and prematurely brushed it aside -- ESPECIALLY when
> addressing the generally public.
Grant V. and Flake R. 1974. "Solutions to the Cost-of-selection
Dilemma". PNAS 71(10) 3863-3865
"This cost-of-selection dilemma has been extensively discussed in the
literature and in informal conferences."
I can only wonder why Walt did not cite this paper in his pulp
fiction....
Walter ReMine
> BTW, I especially like ReMine's claim that Dawkins was helping
> suppress a fact that he might not even have been aware of. (What is
> it with pseudoscientists and their paranoid conspiracy theories,
> anyway?)
I said no such thing.
David Sienkiewicz
Given your past performances, Walter, you'll forgive some of us, I'm sure,
if we don't believe you.
Walter ReMine
>> Indeed, Dawkins's SMALL population size MAGNIFIES THE ILLUSION --
>> for if evolution can proceed this fast (for a single gene, such as
>> hemoglobin, which was the context of Dawkins's discussion), then
>> imagine how EXCEEDINGLY MUCH FASTER the evolution would be in a
>> LARGE population and with MANY genes evolving at this rate! Or so
>> the public was led to believe. Dawkins's program was a whopper
>> illusion-maker, and other evolutionists likewise did little to
>> disabuse the public of it.
>
> I don't know whether anyone "led" the public to believe that or not,
> but FWIW in my own work with genetic algorithms I find that larger
> population sizes almost always help find a better solution within a
> given number of generations. On that basis I would conjecture that
> what you identify as a misleading of the public is in fact correct.
Thanks.
> No, you're the one who's creating the illusion here, by trying to
> interpret Dawkins' program as a model for something Dawkins didn't
> offer it for.
> ....
> IMO it works well as a simple example of how a selection factor can
> make random mutations deliver results _much_ faster ....
Dawkins bragged about its speed, he gave figures for the speed.
Other evolutionists likewise promoted its speed. Therefore, it is entirely
appropriate for me to unmask that speed as an illusion.
I remind you, Dawkins's discussion/program contain two types of illusion:
one about speed, and the other about "success".
If you provide everything needed for success and eliminate everything
preventing success, then the simulation will "succeed". Duh. That is
circular. But evolutionists DON'T POINT OUT those things they are
silenting eliminating -- and that's where the illusion begins. They just
use it to show "success." Whereas I -- by removing some of Dawkins's
silent assumptions that unrealistically favor evolution -- show that the
similation goes into error catastrophe and does not succeed, the target
phrase is NOT reached. The public deserves to know that, and evolutionists
have not told them.
In other words, Dawkins's discussion/program is an illusion-maker, even in
every sense that he uses it.
> I personally would have rather seen [ReMine] use a more conventional
> genetic algorithm rather than the simple WEASEL program, ...
I used Dawkins's WEASEL program because it is easily the most widely known
to the public. It is a powerful illusion-maker, and ripe for debunking. I
show how the ordinary person, equipt with that software (written by
evolutionists themselves), can set it up for less un-natural-ness, and
thereby observe two phenomena that evolutionists (universally?) avoid in
books aimed at the general public:
1) error catastrophe, and
2) a long-term average substitution rate comparable to Haldane's speed
limit.
> .... I don't view most genetic algorithms as very good models
> for nature at all, ...
Evolutionists keep attacking me on the grounds that Dawkins's program is
unrealistic and unnatural. I keep responding that evolutionists are
MISREPRESENTING my material (by omission) -- Dawkins's WEASEL program is
wildly unrealistic IN FAVOR OF EVOLUTION -- and I identify the reasons why.
Calling it 'unrealistic' or 'not a simulation' does not get evolutionists
off the hook.
Also notice that evolutionists have not yet proposed any alternative
genetic algorithm, more realistic than Dawkins's, for such a study. The
reason is that these tend to produce slower speeds than Dawkins's, often
far slower. Moreover, these will still show a key phenomena that I use
Dawkins's program to demonstrate -- that is, other things being equal, when
you lower the reproduction rate this lowers the average long-term
substitution rate. Dawkins's program demonstrates this -- other programs
will too.
>....
> b) Is WEASEL a good way to make that point?
The WEASEL program (and the way it was presented to the public) is a
finely-tuned way to create an illusion. It deserves debunking. Dawkins
was trying to sell evolution to the public, and to do it he ended up
creating a finely-tuned illusion.
> In general smaller populations can harbor less diversity, and
> diversity is absolutely essential for evolution. That's why we
> cosistently see that in genetic algorithms applied to non-trivial
> problems, bigger populations are advantageous.
That explanation is insufficient or incomplete. Genetic diversity does
not help evolution if the genetic "diversity" is comprised of harmful
mutations, as it generally is.
> ....
> You have completely tuned out what Ian has been pointing out for
> three years,
No, for three years Ian Musgrave has been misrepresenting my material
(largely by omission).
> ... you're trying to support your position with a
> conspiracy theory
I said no such thing.
> You claim that WEASEL doesn't address Haldane's Dilemma
No, I said Dawkins didn't address Haldane's Dilemma.
Brian M. Scott
That's exactly what your comments in your 13 August post add up to.
Looks pretty silly when pulled into one place, doesn't it?
Bobby D. Bryant
I guess you'd better file a complaint about someone forging posts in
your name then:
On Tue, 13 Aug 2002 03:05:27 -0600, Walter ReMine wrote:
...
> Dawkins didn't mention Haldane's Dilemma, because evolutionists
> obscured it into oblivion and prematurely brushed it aside --
> ESPECIALLY when addressing the generally public. If Dawkins was
> even aware of Haldane's Dilemma, he had nothing to gain from
> alerting the public to this sleeping dragon.
...
I'm sure this sterling performance is really going to convince a lot
of lurkers that you're a reliable source of information.
Oh, and it's now logged at google for all eternity. Next time
someone asks how reliable you are, I'll be able to direct them
straight to this exchange.
Bobby Bryant
Austin, Texas
Brian M. Scott
<sci...@nospam.net> wrote:
>"Bobby D. Bryant" <bdbr...@mail.utexas.edu> wrote in
>news:ajav5q$ng6$1...@geraldo.cc.utexas.edu:
>>> Indeed, Dawkins's SMALL population size MAGNIFIES THE ILLUSION --
>>> for if evolution can proceed this fast (for a single gene, such as
>>> hemoglobin, which was the context of Dawkins's discussion), then
>>> imagine how EXCEEDINGLY MUCH FASTER the evolution would be in a
>>> LARGE population and with MANY genes evolving at this rate! Or so
>>> the public was led to believe. Dawkins's program was a whopper
>>> illusion-maker, and other evolutionists likewise did little to
>>> disabuse the public of it.
>> I don't know whether anyone "led" the public to believe that or not,
>> but FWIW in my own work with genetic algorithms I find that larger
>> population sizes almost always help find a better solution within a
>> given number of generations. On that basis I would conjecture that
>> what you identify as a misleading of the public is in fact correct.
>Thanks.
Why are you thanking him for telling you that you were wrong? In case
you didn't get it the first time, he's saying that in his experience
this supposed illusion is in fact no illusion at all, but rather very
often the simple truth.
>> No, you're the one who's creating the illusion here, by trying to
>> interpret Dawkins' program as a model for something Dawkins didn't
>> offer it for.
>> ....
>> IMO it works well as a simple example of how a selection factor can
>> make random mutations deliver results _much_ faster ....
>Dawkins bragged about its speed, he gave figures for the speed.
>Other evolutionists likewise promoted its speed. Therefore, it is entirely
>appropriate for me to unmask that speed as an illusion.
Let's restore the part of Bobby's statement that you rather
misleadingly snipped:
>> IMO it works well as a simple example of how a selection factor can
>> make random mutations deliver results _much_ faster than independent
>> random guesses (that ever-popular strawman that stands in for
>> evolution in most creationist arguments).
And here is the point: it works faster than independent random guesses
and so shows the effect of selection. Nothing illusory about that: it
does, and so will any program that applies selection pressure. You
persist in pretending that the program was intended to demonstrate
something more than this.
>I remind you, Dawkins's discussion/program contain two types of illusion:
>one about speed, and the other about "success".
>If you provide everything needed for success and eliminate everything
>preventing success, then the simulation will "succeed". Duh.
How, specifically, does the program 'provide everything needed for
success'?
[...]
>> I personally would have rather seen [ReMine] use a more conventional
>> genetic algorithm rather than the simple WEASEL program, ...
Oh, my, you *are* having trouble reading. The 'him' that you replaced
with '[ReMine]' in fact very obviously referred to Dawkins. Thus,
your comments in response are irrelevant.
[...]
>>....
>> b) Is WEASEL a good way to make that point?
>The WEASEL program (and the way it was presented to the public) is a
>finely-tuned way to create an illusion. It deserves debunking. Dawkins
>was trying to sell evolution to the public, and to do it he ended up
>creating a finely-tuned illusion.
This is gross and blatant dishonesty. Question (b) is meaningless
without the preceding question, which I restore below:
>> a) What point was Dawkins trying to make when he described WEASEL?
You still have not answered this question. It seems rather clear that
you cannot.
>> In general smaller populations can harbor less diversity, and
>> diversity is absolutely essential for evolution. That's why we
>> cosistently see that in genetic algorithms applied to non-trivial
>> problems, bigger populations are advantageous.
>That explanation is insufficient or incomplete. Genetic diversity does
>not help evolution if the genetic "diversity" is comprised of harmful
>mutations, as it generally is.
Are you seriously claiming that genetic diversity in the reproducing
population consists entirely of harmful mutations?
[...]
>> You claim that WEASEL doesn't address Haldane's Dilemma
>No, I said Dawkins didn't address Haldane's Dilemma.
Same thing, unless you're so confused as to think that Haldane's
Dilemma applies to the WEASEL program.
Bobby D. Bryant
Hi Walter. I'll try to be less confrontational in this post,
though I still have to challenge you on several points.
> "Bobby D. Bryant" <bdbr...@mail.utexas.edu> wrote in
> news:ajav5q$ng6$1...@geraldo.cc.utexas.edu:
>
>>> Indeed, Dawkins's SMALL population size MAGNIFIES THE ILLUSION
>>> -- for if evolution can proceed this fast (for a single gene,
>>> such as hemoglobin, which was the context of Dawkins's
>>> discussion), then imagine how EXCEEDINGLY MUCH FASTER the
>>> evolution would be in a LARGE population and with MANY genes
>>> evolving at this rate! Or so the public was led to believe.
>>> Dawkins's program was a whopper illusion-maker, and other
>>> evolutionists likewise did little to disabuse the public of it.
>>
>> I don't know whether anyone "led" the public to believe that or
>> not, but FWIW in my own work with genetic algorithms I find that
>> larger population sizes almost always help find a better solution
>> within a given number of generations. On that basis I would
>> conjecture that what you identify as a misleading of the public
>> is in fact correct.
>
> Thanks.
That response is why I decided to be less confrontational. Let's
see how well this works...
>> No, you're the one who's creating the illusion here, by trying to
>> interpret Dawkins' program as a model for something Dawkins
>> didn't offer it for.
>> ....
>> IMO it works well as a simple example of how a selection factor
>> can make random mutations deliver results _much_ faster ....
>
> Dawkins bragged about its speed, he gave figures for the speed.
> Other evolutionists likewise promoted its speed. Therefore, it is
> entirely appropriate for me to unmask that speed as an illusion.
Sorry, but it sounds to me like you still don't understand Ian's
basic point. (Which really makes me wish you had answered my two
questions below, which you didn't do. I think if you would take
the trouble to answer them we could set 90% of this discussion
aside as an instance of "talking past each other", and then
concentrate on the remaining 10% where the real contention lies.)
[big snip]
>> .... I don't view most genetic algorithms as very good models for
>> nature at all, ...
>
> Evolutionists keep attacking me on the grounds that Dawkins's
> program is unrealistic and unnatural. I keep responding that
> evolutionists are MISREPRESENTING my material (by omission) --
> Dawkins's WEASEL program is wildly unrealistic IN FAVOR OF
> EVOLUTION -- and I identify the reasons why. Calling it
> 'unrealistic' or 'not a simulation' does not get evolutionists off
> the hook.
>
> Also notice that evolutionists have not yet proposed any
> alternative genetic algorithm, more realistic than Dawkins's, for
> such a study. The reason is that these tend to produce slower
> speeds than Dawkins's, often far slower. Moreover, these will
> still show a key phenomena that I use Dawkins's program to
> demonstrate -- that is, other things being equal, when you lower
> the reproduction rate this lowers the average long-term
> substitution rate. Dawkins's program demonstrates this -- other
> programs will too.
>
>>....
>> b) Is WEASEL a good way to make that point?
>
> The WEASEL program (and the way it was presented to the public) is
> a finely-tuned way to create an illusion. It deserves debunking.
> Dawkins was trying to sell evolution to the public, and to do it
> he ended up creating a finely-tuned illusion.
Here is where you should have answered my two questions. You
snipped the first one altogether, and you deflect the second one
with a non-answer. I think it would really help things along if
you would go back and post another reply, answering both questions
clearly. As I said above, I think it would let us ditch about 90%
of this non-discussion and get down to the issues that matter.
>> In general smaller populations can harbor less diversity, and
>> diversity is absolutely essential for evolution. That's why we
>> cosistently see that in genetic algorithms applied to non-trivial
>> problems, bigger populations are advantageous.
>
> That explanation is insufficient or incomplete. Genetic diversity
> does not help evolution if the genetic "diversity" is comprised of
> harmful mutations, as it generally is.
Actually, if I understand correctly most mutations are neutral in
biology.
But either way, it seems that a large pool of mutations is
necessary for a genetic algorithm to make progress. See:
http://www.cs.utexas.edu/users/bdbryant/talk-origins/mutation-experiment/index.html
(Watch out for wrapping link.)
For a GA, since we almost always deploy them to solve a very
specific problem that we can measure the results on, we can
instrument our code and see that mutations do produce a lot of
turkeys alongside the good answers. No problem; we just discard
the turkeys. Yet if we turn off the mutations we get nothing but
turkeys of the worst sort. (See the plots at the link.)
So as I have come to see it, while it may be appropriate to label a
mutation as "good" or "bad" for an individual, it appears that a
population can benefit by having some "bad" mutations in the gene
pool. This is a conjecture, based on the observation of genetic
algorithms rather than any deep knowledge of biology, but it looks
to me like what is "bad" in one generation might turn out to be
"good" later on, either because the environment changed or else
because it got combined with something else that changed its effect.
So I think talking about "bad" mutations is a red herring. I stick
with my claim that large populations are advantageous in GAs
because they can harbor more diversity, and add a new claim that
mutations are necessary for maintaining that diversity. (In GAs,
you breed out all the diversity in the absence of mutations. In
some experiments that I didn't show on the linked page, I found
that the genetic diversity in my GA is apparently a state resulting
from a dynamic equilibrium between the mutation rate and the
tendancy to breed all the diversity out. For example, with lower
mutation rates I still get an approximately flat diversity after
things have settled down, but the lower the mutation rate, the
lower the equilibrium diversity. And of course, as the plot shows,
when you lower the mutation rate to zero you lose _all_ diversity,
and evolution can no longer make any progress.)
It's up for discussion how much of this applies to biology, but I
think it shows at least that we can't appeal to any abstract
principle that says diversity or mutations are bad for evolution.
I think rather that they are absolutely necessary, and that it's
better -- over the long run -- to have a lot of bad mutations than
it is to have no mutations at all.
Remember, evolution isn't looking out for individuals' interests;
it's just a side effect of the way reproduction works. Our goal
isn't to say what would work well vs. what would work poorly; our
goal is just to see what actually happens.
>> ....
>> You have completely tuned out what Ian has been pointing out for
>> three years,
>
> No, for three years Ian Musgrave has been misrepresenting my
> material (largely by omission).
It seems to me that you are completely ignoring his observations.
Instead of responding to them directly you always revert to a
criticism of Dawkins' WEASEL, and Ian's whole point is that those
criticisms are irrelevant.
Again, I would ask you to answer my two questions clearly and
concisely, without throwing in any spurious comments that will
obscure your answers.
>> ... you're trying to support your position with a conspiracy
>> theory
>
> I said no such thing.
Here is what I took to be an accusation of conspiracy:
===8<=======
On Tue, 13 Aug 2002 03:05:27 -0600, Walter ReMine wrote:
...
> Dawkins's program was a whopper illusion-maker, and other
> evolutionists likewise did little to disabuse the public of it.
...
> Dawkins didn't mention Haldane's Dilemma, because evolutionists
> obscured it into oblivion and prematurely brushed it aside --
> ESPECIALLY when addressing the generally public.
...
===8<======
The second sentence in particular makes it sound like you think
scientists, as a group, _deliberately_ led the public down the
wrong path.
>> You claim that WEASEL doesn't address Haldane's Dilemma
>
> No, I said Dawkins didn't address Haldane's Dilemma.
OK, I stand corrected. So try this revision on for size: You said
Dawkins didn't address Haldane's Dilemma; Ian says Dawkins didn't
_try_ to address Haldane's Dilemma. So what's the problem?
If you have something constructive to say about Haldane's Dilemma
then say it. But leave Dawkins out of it, since he didn't address
the issue at all. Or simply say "Dawkins should have addressed
Haldane's Dilemma", and we can discuss that. But as things stand,
it looks like you're just slinging mud, getting defensive when Ian
points out that you're just slinging mud, and then trying to defend
your position by just slinging more mud at the original target,
completely ignoring what Ian pointed out. That sort of "debate"
works in Sunday School, where no one cares to challenge the
speaker, but it doesn't work in front of an audience that expects
people to actually make a case for their claims.
And there are lots of such people in this forum. Plus lots of
lurkers wondering whether a published author can stay in the ring
better than the usual crop of idiots who come here to set us
straight. Show those lurkers that you can answer questions rather
than deflecting them!
And I don't know of any better way for you to start than by answer
the two questions I asked earlier, clearly, concisely, and without
obfuscation.
Bobby Bryant
Austin, Texas
p.s. - Though it's in a different thread, you _really_ ought to
address this while you're here too:
http://groups.google.com/groups?q=haldane%27s+dilemma+group:talk.origins&hl=en&lr=&ie=UTF-8&scoring=d&selm=1513f784.0208141355.5fd385f6%40posting.google.com&rnum=1
(Watch out for the wrapping link.)
Bobby D. Bryant
> On Thu, 15 Aug 2002 05:14:48 +0000 (UTC), Walter ReMine
> <sci...@nospam.net> wrote:
>
>>"Bobby D. Bryant" <bdbr...@mail.utexas.edu> wrote in
>>news:ajav5q$ng6$1...@geraldo.cc.utexas.edu:
>
>>>> Indeed, Dawkins's SMALL population size MAGNIFIES THE ILLUSION
>>>> -- for if evolution can proceed this fast (for a single gene,
>>>> such as hemoglobin, which was the context of Dawkins's
>>>> discussion), then imagine how EXCEEDINGLY MUCH FASTER the
>>>> evolution would be in a LARGE population and with MANY genes
>>>> evolving at this rate! Or so the public was led to believe.
>>>> Dawkins's program was a whopper illusion-maker, and other
>>>> evolutionists likewise did little to disabuse the public of it.
>
>>> I don't know whether anyone "led" the public to believe that or
>>> not, but FWIW in my own work with genetic algorithms I find that
>>> larger population sizes almost always help find a better
>>> solution within a given number of generations. On that basis I
>>> would conjecture that what you identify as a misleading of the
>>> public is in fact correct.
>
>>Thanks.
>
> Why are you thanking him for telling you that you were wrong? In
> case you didn't get it the first time, he's saying that in his
> experience this supposed illusion is in fact no illusion at all,
> but rather very often the simple truth.
LoL. Now that I have re-read the exchange three times, I think he
said "thanks", not for the information (as I also understood him to
be saying), but because he misinterpreted my last sentence as a
vindication of _his_ claim, rather than of the claim that he was
mocking.
I'd be tempted to make a tart comment about creationists and
reading comprehension, if not for:
>>> I personally would have rather seen [ReMine] use a more
>>> conventional genetic algorithm rather than the simple WEASEL
>>> program, ...
>
> Oh, my, you *are* having trouble reading. The 'him' that you
> replaced with '[ReMine]' in fact very obviously referred to
> Dawkins.
Wow... I dislexed right over that one. You're right, of course.
Hopefully he'll take that as an object lesson about about the
dangers of helping the interpretation along with bracketized
substitutions.
Bobby Bryant
Austin, Texas
Roy Thearle
> "Bobby D. Bryant" <bdbr...@mail.utexas.edu> wrote in
> news:ajav5q$ng6$1...@geraldo.cc.utexas.edu:
> > I don't know whether anyone "led" the public to believe that or not,
> > but FWIW in my own work with genetic algorithms I find that larger
> > population sizes almost always help find a better solution within a
> > given number of generations. On that basis I would conjecture that
> > what you identify as a misleading of the public is in fact correct.
>
> Thanks.
Huh? Did you actually understand what he wrote?
[deletia]
> > I personally would have rather seen [ReMine] use a more conventional
> > genetic algorithm rather than the simple WEASEL program, ...
>
> I used Dawkins's WEASEL program because it is easily the most widely known
> to the public. It is a powerful illusion-maker, and ripe for debunking. I
> show how the ordinary person, equipt with that software (written by
> evolutionists themselves), can set it up for less un-natural-ness,
Do you really think setting a maximum population size of 6 is making
the simulation _less_ unnatural?
> Evolutionists keep attacking me on the grounds that Dawkins's program is
> unrealistic and unnatural.
No. Evolutionists keep attacking you because you
(i) misrepresent them
(ii) do not apparently know what you are doing.
[deletia II]
> Moreover, these will still show a key phenomena that I use
> Dawkins's program to demonstrate -- that is, other things being equal, when
> you lower the reproduction rate this lowers the average long-term
> substitution rate. Dawkins's program demonstrates this -- other programs
> will too.
Since in these programs the reproduction rate is linked to the
population
size, you _cannot_ change the reproduction rate while having other
things
being equal. Do you understand this?
[deletia III - the sequel]
> > In general smaller populations can harbor less diversity, and
> > diversity is absolutely essential for evolution. That's why we
> > cosistently see that in genetic algorithms applied to non-trivial
> > problems, bigger populations are advantageous.
>
> That explanation is insufficient or incomplete. Genetic diversity does
> not help evolution if the genetic "diversity" is comprised of harmful
> mutations, as it generally is.
Is it? Aren't most mutations neutral?
> > ....
> > You have completely tuned out what Ian has been pointing out for
> > three years,
>
> No, for three years Ian Musgrave has been misrepresenting my material
> (largely by omission).
And for three years you have been misrepresenting him by deliberate
misquote.
Are you going to acknowledge and/or correct this?
[son of deletia rides again]
Roy
Adam Marczyk
news:1513f784.02081...@posting.google.com...
Ah, but you see, because "evolutionists" didn't rent huge billboards on every
major highway in America where they announced in big flashing neon letters, "WE
HAVE A PROBLEM CALLED HALDANE'S DILEMMA!", what they did actually qualifies as
hiding it. (We know our literature and conferences have too many big words for
ordinary people to even *try* understanding them!)
BTW, does anyone else find incredible the sheer arrogance of ReMine's claim that
scientists did everything in their power to obscure Haldane's Dilemma until *he*
wrote about it and forced them to bring it to light? You almost have to admire
the chutzpah to make a claim like that, especially when it's so easy to show how
blazingly and utterly false it is, as several people have done.
Adam Marczyk
> "Bobby D. Bryant" <bdbr...@mail.utexas.edu> wrote in
> news:ajav5q$ng6$1...@geraldo.cc.utexas.edu:
>
> >> Indeed, Dawkins's SMALL population size MAGNIFIES THE ILLUSION --
> >> for if evolution can proceed this fast (for a single gene, such as
> >> hemoglobin, which was the context of Dawkins's discussion), then
> >> imagine how EXCEEDINGLY MUCH FASTER the evolution would be in a
> >> LARGE population and with MANY genes evolving at this rate! Or so
> >> the public was led to believe. Dawkins's program was a whopper
> >> illusion-maker, and other evolutionists likewise did little to
> >> disabuse the public of it.
> >
> > I don't know whether anyone "led" the public to believe that or not,
> > but FWIW in my own work with genetic algorithms I find that larger
> > population sizes almost always help find a better solution within a
> > given number of generations. On that basis I would conjecture that
> > what you identify as a misleading of the public is in fact correct.
>
> Thanks.
I think you misread him. He said that Dawkins' simulation was correct, not your
identifying it as misleading.
> > No, you're the one who's creating the illusion here, by trying to
> > interpret Dawkins' program as a model for something Dawkins didn't
> > offer it for.
> > ....
> > IMO it works well as a simple example of how a selection factor can
> > make random mutations deliver results _much_ faster ....
>
> Dawkins bragged about its speed, he gave figures for the speed.
> Other evolutionists likewise promoted its speed. Therefore, it is entirely
> appropriate for me to unmask that speed as an illusion.
>
> I remind you, Dawkins's discussion/program contain two types of illusion:
> one about speed, and the other about "success".
>
> If you provide everything needed for success and eliminate everything
> preventing success, then the simulation will "succeed". Duh. That is
> circular. But evolutionists DON'T POINT OUT those things they are
> silenting eliminating -- and that's where the illusion begins. They just
> use it to show "success." Whereas I -- by removing some of Dawkins's
> silent assumptions that unrealistically favor evolution -- show that the
> similation goes into error catastrophe and does not succeed, the target
> phrase is NOT reached. The public deserves to know that, and evolutionists
> have not told them.
Dawkins' weasel program is *not* intended to simulate evolution. It was *never*
claimed to do that. Its only purpose is to show the superiority of cumulative
multi-step selection over single-step random guessing. I don't know whether your
repeated failure to grasp this simple point is deliberate or not, but
steadfastly refusing to acknowledge it is not going to do much to inspire
confidence in the quality of your argument.
> In other words, Dawkins's discussion/program is an illusion-maker, even in
> every sense that he uses it.
>
> > I personally would have rather seen [ReMine] use a more conventional
> > genetic algorithm rather than the simple WEASEL program, ...
>
> I used Dawkins's WEASEL program because it is easily the most widely known
> to the public. It is a powerful illusion-maker, and ripe for debunking. I
> show how the ordinary person, equipt with that software (written by
> evolutionists themselves), can set it up for less un-natural-ness, and
> thereby observe two phenomena that evolutionists (universally?) avoid in
> books aimed at the general public:
>
> 1) error catastrophe, and
> 2) a long-term average substitution rate comparable to Haldane's speed
> limit.
>
> > .... I don't view most genetic algorithms as very good models
> > for nature at all, ...
>
> Evolutionists keep attacking me on the grounds that Dawkins's program is
> unrealistic and unnatural. I keep responding that evolutionists are
> MISREPRESENTING my material (by omission) -- Dawkins's WEASEL program is
> wildly unrealistic IN FAVOR OF EVOLUTION -- and I identify the reasons why.
> Calling it 'unrealistic' or 'not a simulation' does not get evolutionists
> off the hook.
Again, Dawkins' program is not intended to simulate evolution, merely to
demonstrate the power of cumulative selection over random single-step guessing.
You are the one who is blatantly misrepresenting and failing to understand the
things you claim to criticize.
> Also notice that evolutionists have not yet proposed any alternative
> genetic algorithm, more realistic than Dawkins's, for such a study. The
> reason is that these tend to produce slower speeds than Dawkins's, often
> far slower. Moreover, these will still show a key phenomena that I use
> Dawkins's program to demonstrate -- that is, other things being equal, when
> you lower the reproduction rate this lowers the average long-term
> substitution rate. Dawkins's program demonstrates this -- other programs
> will too.
>
> >....
> > b) Is WEASEL a good way to make that point?
>
> The WEASEL program (and the way it was presented to the public) is a
> finely-tuned way to create an illusion. It deserves debunking. Dawkins
> was trying to sell evolution to the public, and to do it he ended up
> creating a finely-tuned illusion.
Note you snipped without marking the first question - what point was the weasel
program intended to demonstrate? Since answering this question would have most
clearly displayed your misunderstanding, it's no surprise you silently deleted
it. I can only conclude that, if your ignorance is not intentional, it's powered
by cognitive dissonance of a most subtle and nefarious sort.
Noctiluca
> On Wed, 14 Aug 2002 23:14:48 -0600, Walter ReMine wrote:
>
> And I don't know of any better way for you to start than by answer
> the two questions I asked earlier, clearly, concisely, and without
> obfuscation.
If he does this honestly, based on the obvious reality of Dawkins'
intent, then the force of his falsely-premised argument is lost.
Reversing his position on this issue would actually be a rather small
matter within the larger context of the ID-Evo debate. However, for
those who consider their mission to be of holy provenance even an
insignificant concession is often unacceptable.
I await the results with bated, if not held, breath.
robert
James Acker
: BTW, does anyone else find incredible the sheer arrogance of ReMine's claim
: that
: scientists did everything in their power to obscure Haldane's Dilemma
: until *he*
: wrote about it and forced them to bring it to light? You almost have to
: admire
: the chutzpah to make a claim like that, especially when it's so easy to
: show how
: blazingly and utterly false it is, as several people have done.
Back in April, in the thread entitled "Kooks, Nuts, Cranks, etc.",
I wrote about the Discover magazine article "Notes from a Parallel
Universe" describing scientific kooks, nuts, cranks, etc. One of the classic
hallmarks of a crackpot is the utter conviction that they (and they
*alone*) have discovered a fundamental misunderstanding at the core of a
widely-held (and famous) scientific theory. Many of these crackpots
will go to extraordinary lengths -- including the self-publication of
a large and extensive book -- to publicize their supposed discovery
and to bring it to the attention of the unknowing masses of humanity.
ReMine seems too calculating and devious to qualify as a true
crackpot, but some aspects of his pursuits share the characteristics
of crackpot artistry. One of the aspects of crackpottery he shares
is the utter inability to understand any corrections to the fundamental
errors of understanding which lead the crackpot to believe that they
have discovered a major flaw in a scientific theory.
Ian Musgrave & Peta O'Donohue
Address altered to avoid spam, delete RemoveInsert
On Thu, 15 Aug 2002 05:14:48 +0000 (UTC), Walter ReMine
<sci...@nospam.net> wrote:
>"Bobby D. Bryant" <bdbr...@mail.utexas.edu> wrote in
>news:ajav5q$ng6$1...@geraldo.cc.utexas.edu:
[snip]
>> I personally would have rather seen [ReMine] use a more conventional
>> genetic algorithm rather than the simple WEASEL program, ...
Mr ReMine, please read for comprehension, the "he" that you replaced
with [ReMine] was _Dawkins_.
>I used Dawkins's WEASEL program because it is easily the most widely known
>to the public. It is a powerful illusion-maker, and ripe for debunking. I
>show how the ordinary person, equipt with that software (written by
>evolutionists themselves), can set it up for less un-natural-ness, and
>thereby observe two phenomena that evolutionists (universally?) avoid in
>books aimed at the general public:
>
>1) error catastrophe, and
>2) a long-term average substitution rate comparable to Haldane's speed
>limit.
The problem is that you do not understand either Dawkins program or
Haldane's dilemma.
1) In Dawkins program the substitution rate is CONSTANT, and is 1.
That is _any_ benefical mutation(s) whether one or more, will be fixed
in one generation.
2) Haldane's Dilemma is not "evolution has a speed limit". Haldane's
dilemma is that you can only fix two or more simultaneous mutations at
the same rate as two or more sequential mutations (under certain
conditions).
That you can manipulate Dawkins program so that it only runs twice as
fast as Haldane's average estimate for long gestation vertebrates
under certain assumptions (independence of mutations, low selection
co-efficents, hard rather than soft selection, no gene segregation
etc.) says nothing about either Haldane's dilemma or evolution, but
speaks volumes about your profound ignorance of both.
[snip]
Walter ReMine
news:ulnp71...@corp.supernews.com:
> .... ReMine's claim that scientists did everything
> in their power to obscure Haldane's Dilemma ...
Adam Marczyk
Mr. ReMine, really. If you want to lie about your own words, fine, but generally
it's considered good Usenet arguing form to not do so as a reply to a post that
actually contained the words you wanted to deny ever having said. Do I really
have to go to Google here?
http://groups.google.com/groups?selm=Xns92692A9E838CEsciencenospamnet%40206.191.
138.115
"Dawkins didn't mention Haldane's Dilemma, because evolutionists obscured it
into oblivion and prematurely brushed it aside -- ESPECIALLY when
addressing the generally public. If Dawkins was even aware of Haldane's
Dilemma, he had nothing to gain from alerting the public to this sleeping
dragon. The Haldane-based limitation of approximately 1667 beneficial
nucleotides for human evolution was not something evolutionists wanted
taught publicly -- and indeed they never did (until after my book)."
--
Ian Musgrave & Peta O'Donohue
Address altered to avoid spam, delete RemoveInsert
On Thu, 15 Aug 2002 05:14:48 +0000 (UTC), Walter ReMine
<sci...@nospam.net> wrote:
>"Bobby D. Bryant" <bdbr...@mail.utexas.edu> wrote in
>news:ajav5q$ng6$1...@geraldo.cc.utexas.edu:
[snip]
>> ....
>> You have completely tuned out what Ian has been pointing out for
>> three years,
>
>No, for three years Ian Musgrave has been misrepresenting my material
>(largely by omission).
It may suprise Mr. ReMine, but scientists tend to concentrate on a key
question, rather than trying to answer a whole lot of questions at the
same time.
In the case in point, the question is whether Mr. Remine's claim that
Dawkins' weasel program (or implementations of it) demonstrates
Haldane's dilemma is correct. This is a key question, and many of Mr.
ReMine's ancillary claims hang off this point.
Addressing this key question in detail is not misrepresenting Mr.
ReMines argument.
As it so happens, Dawkins' program does not, cannot, demonstrate
Haldane's dilemma (because the substitution rate is a constant one
generation), which significantly impacts on the rest of Mr. ReMines
arguments.
Brian M. Scott
--- begin quote ---
From: Walter ReMine <sci...@nospam.net>
Newsgroups: talk.origins
Subject: Re: Weasel program -- How evolutionists misrepresent ReMine
Date: Tue, 13 Aug 2002 09:05:27 +0000 (UTC)
Organization: University of Ediacara
Lines: 136
Sender: ro...@darwin.ediacara.org
Approved: rob...@ediacara.org
Message-ID: <Xns92692A9E838CE...@206.191.138.115>
[...]
Dawkins didn't mention Haldane's Dilemma, because evolutionists
obscured it into oblivion and prematurely brushed it aside --
ESPECIALLY when addressing the generally public. If Dawkins was even
aware of Haldane's Dilemma, he had nothing to gain from alerting the
public to this sleeping dragon. The Haldane-based limitation of
approximately 1667 beneficial nucleotides for human evolution was not
something evolutionists wanted taught publicly -- and indeed they
never did (until after my book).
[...]
--- end quote ---
Bobby D. Bryant
> Walter ReMine <sci...@nospam.net> wrote in message
> news:Xns926BDBA52596C...@206.191.138.115...
>> "Adam Marczyk" <ebon...@hotmailNOTexcite.com> wrote in
>> news:ulnp71...@corp.supernews.com:
>>
>> > .... ReMine's claim that scientists did everything in their
>> > power to obscure Haldane's Dilemma ...
>>
>> I said no such thing.
>
> Mr. ReMine, really. If you want to lie about your own words, fine,
> but generally it's considered good Usenet arguing form to not do
> so as a reply to a post that actually contained the words you
> wanted to deny ever having said.
Yes, you're supposed to wait a couple of weeks and hope that no one
remembers or cares enough to visit google to check.
But the funny thing is, "I said no such thing" has been so
consistently false in Mr. ReMine's posts that you can safely take
it as a sign that he _did_ say it.
Ironic, coming from someone who inserts bracketized revisions into
other people's posts and excuses them with "hardly changed the
meaning".
Bobby Bryant
Austin, Texas
Roy Thearle
But you did claim this:
"... evolutionists obscured it [Haldane's dilemma]
into oblivion and prematurely
brushed it aside -- ESPECIALLY when
addressing the generally [sic] public."
So effectively your claim is that scientists obscured Haldane's dilemma -
but they weren't trying to.
Correct?
Roy
David Wilson
in talk.origins rthe...@hotmail.com (Roy Thearle) wrote:
> ... I've been trying to remember what Dawkins
> originally wrote in "The Blind Watchmaker", as I won't have access to
> my copy for a while.
>
> Is it correct that the original program didn't have a population at
> all, but just had a single string for which every letter that didn't
> match the target was mutated in every generation?
>
Here is Dawkins's description (from the Penguin edition):
"It [the cumulative selection procedure--djw] again begins by choosing
a random sequence of 28 letters, just as before [i.e. just as in the
case of single-step selection---djw]:
WDLMNLT DTJBKWIRZREZLMQCO P
[sic---it is evident from the next generation of the string given by
Dawkins that there is a letter missing---probably a 'T'---from the
fourth position of this string, between the 'L' and the 'M']
It now 'breeds from' this random phrase. It duplicates repeatedly
but with a certain chance of random error--'mutation'--in the
copying. The computer examines the mutant nonsense phrases, the
'progeny' of the original phrase, and chooses the one which, _however
slightly_, most resembles the target phrase, METHINKS IT IS LIKE A
WEASEL. In this instance the winning phrase of the next 'generation'
happened to be:
WDLTMNLT DTJBSWIRZREZLMQCO P
Not an obvious improvement! But the procedure is repeated, again
mutant 'progeny' are 'bred from' the phrase, and a new 'winner' is
chosen. This goes on, generation after generation."
I can't find anything in Dawkins's description to indicate that _only_
letters not matching those in the corresponding positions of the target
string were subjected to 'mutation'. William Dembski has claimed that
this was so, but I have never found any evidence to support that claim.
Nor can I find anything in Dawkins's description to indicate that he used
exactly one mutation per progeny, as claimed by Walter Remine.
Dawkins also doesn't seem to say anywhere in "The Blind Watchmaker" how
many offspring each parent string produced in his program, or even if this
number was kept constant from one generation to the next. While I am
unaware of anything that would specifically contradict Mr Remine's claim
that Dawkins used 100 offspring per parent, I have also seen no evidence
that would justify it either.
Mr Remine has acknowledged that his analysis was carried out using a program
called MONKEY, written by David Wise, and available from Mr Wise's home site
at:
http://members.aol.com/dwise1/cre_ev/monkey.html
The options and parameter values which Mr Remine claims that Dawkin's used
are (apart from the target phrase) simply the default values assigned to
them by this program of Mr Wise's. It's not at all clear (at least not to
me) from Dawkins's description in "The Blind Watchmaker" that the procedure
he used was identical to any of those implemented in Mr Wise's program. It
_might_ have been, but Dawkins doesn't seem to me to have given sufficiently
precise details in his book for anyone to be able to tell for sure.
Apart from the description quoted above, the only information Dawkins gives
that could throw more light on the details of the procedure he used are a
sequence of outputs from his program and the numbers of generations which
three runs of it took to reach the target. These were 43, 64 and 41.
When Wise's program is run with Dawkins's target phrase "METHINKS IT IS
LIKE A WEASEL", and other options having the default values assigned
to them by the program, the expected number of generations for the process
to reach the target string is 46.59 with a standard deviation of 9.56.
While this is reasonably consistent with the figures Dawkin's gives, so is
a wide range of other combinations of options and parameter values.
----------------------------------------------------------------------------
David Wilson
SPAMMERS_fingers@SHOULD_BE_funnelweb_PROSECUTED.com.au
(Remove underlines and upper case letters to obtain my email address)
Mitchell Coffey
Adam wrote: "BTW, does anyone else find incredible the sheer arrogance
of ReMine's claim that scientists did everything in their power to
obscure Haldane's Dilemma until *he* wrote about it and forced them to
bring it to light?"
> I said no such thing.
What you wrote was this, your response when someone pointed out that
Dawkins never made any claims regarding his Weasle program and
Haldane's dilemma:
"Dawkins didn't mention Haldane's Dilemma, because evolutionists
obscured it
into oblivion and prematurely brushed it aside -- ESPECIALLY when
addressing the generally public."
It is ceratainly possible that evolutionists were able to obscure
Haldane's dilemma into oblivion using only a limited subset of their
powers. Adam's mild hyperbolie is well within acceptable bounds,
however - ask your colleagues at Discovery. You, however, continue to
obscure the fact that your on going complaint about Dawkins, the
Weasle program and Haldane's dilemma is without merit. You're
attributing prevarication to Dawkins and "evolutionists" in general,
and yet you can't even produce a quote from Dawkins supporting you!
Mitchell Coffey
David Sienkiewicz
"No such thing?"
It's quite obvious that you did, Walter.
catshark
I confess that I have not paid much attention to your previous forays
into t.o., so I am genuinely confused by this. For a second time in
this thread you have made the bald assertion that you "said no such
thing", only to have the posts quoted back where apparently you made
the exact statement or close enough. If you are claiming that those
quoting you are doing it out of context or misinterpreting it, why
then is there no attempt to elucidate? How can anyone coming "cold"
to your arguments conclude other than:
1) You are oblivious to what you yourself write, or
2) You are dishonest *and* oblivious to the existence of
Google.com, or
3) You somehow think that playing the fool enhances your
credibility?
---------------
J. Pieret
---------------
Only two things are infinite,
the universe and human stupidity,
and I'm not sure about the former.
- Albert Einstein -
David Wilson
in talk.origins Walter ReMine <sci...@nospam.net> wrote:
> ... I -- by removing some of Dawkins's
> silent assumptions that unrealistically favor evolution -- show that the
> similation goes into error catastrophe and does not succeed, the target
> phrase is NOT reached. ...
This is nonsense. Mr Remine has acknowledged that he used David Wise's
program MONKEY to carry out his experiments. The only reason why his
attempts to run this program would not have succeeded is that he did not
allow it to run long enough for it to reach the target.
In the procedures implemented by Mr Wise's program the numbers of positions
at which the target string matches the parent strings of successive
generations can be regarded as successive states in a process called an
"irreducible Markov chain". If allowed to keep running, such a process will
eventually reach _any_ of its states with probability 1.
It is true that by making the mutation rate of the characters in the string
sufficiently large, the expected waiting time for Mr Wise's program to reach
the target string can be made so large that one would not expect it to
terminate within the lifetime of the universe. However, the only way
one can tell whether some particular value of the mutation rate is large
enough to have this effect is to use a reliable method for calculating
the expected waiting time, or at least a lower bound for it. Monte
Carlo experiments which are terminated before the target string is
reached are useless for this purpose. All they can tell you is that
the expected waiting time for the target to be reached is likely to be
somewhat longer than the time you allowed your experiments to run.
But Mr Remine apparently bases his claim on the results of just such Monte
Carlo experiments. This would give him no justification whatever for
asserting that the program "does not succeed". In fact, for the options and
parameters he claims to have used in his supposed demonstration of "error
catastrophe", the assertion that the program does not succeed is ludicrous.
The fullest description I have seen of these options and parameters is
contained in quotations from Mr Remine's book, "The Biotic Message", by
Ian Musgrave (URL http://makeashorterlink.com/?X1E732D71 --- I have taken
the liberty of correcting a couple of obvious typos):
> <begin quotes>
> p 235
>
> "That method of mutation is not true to nature [used by Dawkins]. In
> nature nothing counts mutations and assures exactly one in each
> progeny. A more realistic type of mutation should be used in the
> simulation so that each letter has a _probability_ of mutation.
> Suppose we use this correct method of mutation while leaving the
> "average" rate unchanged (at 1 chance in 28).
> This subtle correction to the simulation nearly doubles the time
> needed to evolve the target phrase: to 86 generations."
>
> p 236
>
> "Then we reduce the reproduction rate to that of the higher
> vertebrates, say n=6. In a sexual species this would require the
> females to produce 12 offspring each. This is overly optimistic for
> many species. The simulation then goes into error catastrophe and does
> not reach the target phrase. ...
>
For these options and parameters, the expected waiting time for MONKEY to
reach the target string turns out to be 348,048 generations (to the nearest
integer). This is _not_ a Monte Carlo estimate. It is the nearest integer
to the _exact_ expected waiting time for the above-mentioned Markov chain
to enter its absorbing state (i.e. the target state).
On my aging Pentium III home computer, Mr Wise's program will run through
about 200,000 generations per minute, so there is no problem whatever running
it with the parameter values given above until it reaches the target. A few
days ago I did this 10 times with the following results:
Run #Generations Run #Generations
to hit target to hit target
1. 650,253 6. 21,566
2. 326,900 7. 856,444
3. 181,592 8. 81,995
4. 1,064,229 9. 1,678,545
5. 155,536 10. 236,424
To forestall any accusations by Mr Remine that I must have used an incorrect
option when running the program, I include a copy of the input screen
below:
1. Target: User Entered --
"METHINKS IT IS LIKE A WEASEL"
2. Option: Cumulative Selection -- All Offspring Change
3. Method: Chance of Changing Any or All Letters
4. Number of Offspring: 6
5. Base Probability of Change: 0.0357
David Wilson
in talk.origins "Ian Musgrave & Peta O'Donohue"
<reynella_R...@werple.mira.net.au> wrote:
> YOU [Walter Remine--djw] have claimed,
> repeatedly, that Dawkins program demonstrates Haldane's dilemma.
> ...
As you are aware, the program Mr Remine used was actually written by
David Wise. The one Dawkins wrote may well have differed from it
(though probably in relatively unimportant ways). By continually
attributing this program and its default parameters to Dawkins, Mr Remine
both misprepresents Dawkins and deprives Wise of the proper credit due to
him.
> Haldane's Dilemma is NOT "evolution goes slow", it is that two or more
> simultaneous beneficial mutations will take as long to fix in a
> population as two or more sequential beneficial mutations. In Dawkins
> program beneficial mutations, no matter how many, are _always_ fixed
> in one generation. ...
While this is true, it is unfortunately obscured from anyone who doesn't
understand some of the details of the program's operation by the long
time that elapses before the target character appears in each position.
of the evolving string. If the "fitnesses" of the substituting letters
at different positions on the evolving string were independent (which,
of course, they aren't), Haldane's argument would establish that the
target string could almost never be reached in fewer than 28 generations.
While Haldane's argument is inapplicable to the procedures implemented
in Wise's program, the conclusion that the target will hardly ever be
reached in fewer that 28 generations seems to be correct. For example,
with Dawkins's target string "METHINKS IT IS LIKE A WEASEL", and the
default values for the other options and parameters of the program,
the probability that it will reach the target in 28 or fewer generations
is about 7.9 x 10^(-4). This will therefore occur on only about one in
every 1,250 executions of the program with those parameters.
The failure of Haldane's argument _for this type of process_ can be
illustrated spectacularly, however, by reducing the alphabet size.
With an alphabet size of 2, target string of 28 B's, starting string
of 28 A's (say), mutation rate of 0.2 per character, and 100 offspring
per parent, the expected waiting time for the target to be reached is
only 10.4 generations, much less than the 28 generations it would have
to _at least_ be if Haldane's argument were applicable. There is a
also a 99.9% chance that the target will be reached on or before the
21st generation.
Eric Gill
news:273ab496.02081...@posting.google.com:
> Walter ReMine <sci...@nospam.net> wrote in message
> news:<Xns926BDBA52596C...@206.191.138.115>...
>> "Adam Marczyk" <ebon...@hotmailNOTexcite.com> wrote in
>> news:ulnp71...@corp.supernews.com:
>>
>>> .... ReMine's claim that scientists did everything
>>> in their power to obscure Haldane's Dilemma ...
>>
>>I said no such thing.
>
> I confess that I have not paid much attention to your previous forays
> into t.o., so I am genuinely confused by this. For a second time in
> this thread you have made the bald assertion that you "said no such
> thing", only to have the posts quoted back where apparently you made
> the exact statement or close enough. If you are claiming that those
> quoting you are doing it out of context or misinterpreting it, why
> then is there no attempt to elucidate?
This is purest ReMine. If you feel like wasting the time, a Google search
will turn up many other occasions of this irrational yet distinct behavior,
especially on threads discussing his last book.
<snip>
Noelie S. Alito
news:273ab496.02081...@posting.google.com...
> Walter ReMine <sci...@nospam.net> wrote in message
news:<Xns926BDBA52596C...@206.191.138.115>...
> > "Adam Marczyk" <ebon...@hotmailNOTexcite.com> wrote in
> > news:ulnp71...@corp.supernews.com:
> >
> >> .... ReMine's claim that scientists did everything
> >> in their power to obscure Haldane's Dilemma ...
> >
> >I said no such thing.
>
> I confess that I have not paid much attention to your previous forays
> into t.o., so I am genuinely confused by this. For a second time in
> this thread you have made the bald assertion that you "said no such
> thing", only to have the posts quoted back where apparently you made
> the exact statement or close enough.
<cough> He *typed* it.
> If you are claiming that those
> quoting you are doing it out of context or misinterpreting it, why
> then is there no attempt to elucidate? How can anyone coming "cold"
> to your arguments conclude other than:
>
> 1) You are oblivious to what you yourself write, or
> 2) You are dishonest *and* oblivious to the existence of
> Google.com, or
> 3) You somehow think that playing the fool enhances your
> credibility?
4) His brain invokes Reset/Reload-from-archive after every post.
Noelie
--
If you argue with an obsessive nutjob, chances are he is doing the same.
Roy Thearle
> In article <68494300.02081...@posting.google.com> on August 13th
> in talk.origins rthe...@hotmail.com (Roy Thearle) wrote:
>
> > ... I've been trying to remember what Dawkins
> > originally wrote in "The Blind Watchmaker", as I won't have access to
> > my copy for a while.
> >
> > Is it correct that the original program didn't have a population at
> > all, but just had a single string for which every letter that didn't
> > match the target was mutated in every generation?
> >
>
> Here is Dawkins's description (from the Penguin edition):
Many thanks.
Roy
catshark
> "catshark" <catsh...@yahoo.com> wrote in message
> news:273ab496.02081...@posting.google.com...
> > Walter ReMine <sci...@nospam.net> wrote in message
> news:<Xns926BDBA52596C...@206.191.138.115>...
> > > "Adam Marczyk" <ebon...@hotmailNOTexcite.com> wrote in
> > > news:ulnp71...@corp.supernews.com:
> > >
> > >> .... ReMine's claim that scientists did everything
> > >> in their power to obscure Haldane's Dilemma ...
> > >
> > >I said no such thing.
> >
> > I confess that I have not paid much attention to your previous forays
> > into t.o., so I am genuinely confused by this. For a second time in
> > this thread you have made the bald assertion that you "said no such
> > thing", only to have the posts quoted back where apparently you made
> > the exact statement or close enough.
>
> <cough> He *typed* it.
Ah!
>
> > If you are claiming that those
> > quoting you are doing it out of context or misinterpreting it, why
> > then is there no attempt to elucidate? How can anyone coming "cold"
> > to your arguments conclude other than:
> >
> > 1) You are oblivious to what you yourself write, or
> > 2) You are dishonest *and* oblivious to the existence of
> > Google.com, or
> > 3) You somehow think that playing the fool enhances your
> > credibility?
>
> 4) His brain invokes Reset/Reload-from-archive after every post.
There is an assertion without evidence in there.
---------------
J. Pieret
---------------
The secret of success is honesty and fair dealing.
If you can fake those, you've got it made.
- Groucho Marx -
Walter ReMine
> ....
> Nor can I find anything in Dawkins's description
> to indicate that he used exactly one mutation per progeny, as claimed
> by Walter ReMine.
> ....
> While I am unaware of anything that would specifically contradict Mr
> Remine's claim that Dawkins used 100 offspring per parent, I have also
> seen no evidence that would justify it either.
Wilson suggests far more confusion than is warranted by the evidence. As
stated in my book, merely change the simulation's unrealistic deterministic
mutation (of EXACTLY one mutation per progeny) to a more realistic NON-
deterministic mutation (while keeping the mutation RATE the same as before,
and reproduction rate N=100) -- this nearly DOUBLES the time needed to
reach the target phrase.
In other words, Dawkins used a reproduction rate, N, of **AT LEAST** 100!
(In a sexual species that would require females to average 200 progeny
each.)
> Dawkins also doesn't seem to say anywhere in "The Blind Watchmaker"
> how many offspring each parent string produced in his program, ....
That's correct -- Dawkins didn't tell the public about the high
reproduction rate (N >= 100) that he used. AND THAT MADE THE EVOLUTIONARY
ILLUSION MORE EFFECTIVE! Dawkins's simulation contained many, many
unrealistic factors IN FAVOR OF EVOLUTION, yet Dawkins did not draw
attention to them. Instead he focused on its "success" and its speed. The
illusion was nearly perfect.
> Mr ReMine has acknowledged that his analysis was carried out using a
> program called MONKEY, written by David Wise, ...
That acknowledgement is in my book, along with where to obtain a copy of
the simulation. To begin my research, I had first sent Dawkins a check,
requesting a copy of his program. He kindly wrote back, and returned the
check, saying the simulation had been "lost." The fact that Dawkins's
program was lost is unfortunate, but -- if push comes to shove -- it is yet
another reason to publicly denigrate Dawkins's discussion of it. David
Wise's program is the evolutionists' own best reconstruction of Dawkins's
program, and reproduces Dawkins's published results exactly. So long as
Dawkins's simulation retains any luster in the public mind, it will remain
the target of my criticism. Evolutionists' attempts to divert this to a
discussion of "David Wise's program" is, in my view, an attempt to allow
Dawkins's illusions to continue.
-- Walter ReMine
Fellow with Discovery Institute
Center for the Renewal of Science and Culture
_The Biotic Message_
http://www1.minn.net/~science
Ian Musgrave & Peta O'Donohue
Address altered to avoid spam, delete RemoveInsert
On Mon, 19 Aug 2002 07:07:03 +0000 (UTC), Walter ReMine
<sci...@nospam.net> wrote:
>David Wilson <see_sig@for_my.address> wrote in
>news:200208161436...@funnelwebinternet.com.au:
>
>> ....
>> Nor can I find anything in Dawkins's description
>> to indicate that he used exactly one mutation per progeny, as claimed
>> by Walter ReMine.
>> ....
>> While I am unaware of anything that would specifically contradict Mr
>> Remine's claim that Dawkins used 100 offspring per parent, I have also
>> seen no evidence that would justify it either.
>
>Wilson suggests far more confusion than is warranted by the evidence. As
>stated in my book, merely change the simulation's unrealistic deterministic
>mutation (of EXACTLY one mutation per progeny) to a more realistic NON-
>deterministic mutation (while keeping the mutation RATE the same as before,
>and reproduction rate N=100) -- this nearly DOUBLES the time needed to
>reach the target phrase.
However, the whole point is that there is no evidence that Dawkins
_did_ use exactly one mutation per progeny.
Dawkins program was simply contrasting a toy mutation and selection
system with a "typing monkey" random string generator. The point being
that mutation and selection was much, much faster by many orders of
magnitude than random search. Even if you slow implementations of
Dawkins program down a thousand fold, it is still wildly better than
random search (a few minutes on a modern computer for selection vs
around a billion years on a modern computer for random search) Dawkins
point is the same, mutation and selection are staggeringly faster than
random search.
While Mr. ReMine changed a perceived "unrealistic" parameter which
slowed the program down, he did not change another unrealistic
parameter, the symbol set is 27 characters (26 english letters and a
space) whereas in real organisms the symbol set is 4 characters
(ATGC), which would make the program go _much_ faster (even if you
used the protein symbol set of 20 characters it would still go
significantly faster).
>In other words, Dawkins used a reproduction rate, N, of **AT LEAST** 100!
>(In a sexual species that would require females to average 200 progeny
>each.)
In Dawkins program, only _one_ string is copied.
>> Dawkins also doesn't seem to say anywhere in "The Blind Watchmaker"
>> how many offspring each parent string produced in his program, ....
>
>That's correct -- Dawkins didn't tell the public about the high
>reproduction rate (N >= 100) that he used. AND THAT MADE THE EVOLUTIONARY
>ILLUSION MORE EFFECTIVE! Dawkins's simulation contained many, many
>unrealistic factors IN FAVOR OF EVOLUTION, yet Dawkins did not draw
>attention to them.
No, he drew attention to the _main_ reason that the program did not
mimic natural selection in the wild (and pointed out that it was not
the only reason either).
>Instead he focused on its "success" and its speed. The
>illusion was nearly perfect.
No, he pointed out that while the program illustrated the efficacy of
selection over random search, it was not a good illustration of
natural selection in the wild, then discussed a program that would
simulate natural selection in the wild.
James Acker
: While Mr. ReMine changed a perceived "unrealistic" parameter which
: slowed the program down, he did not change another unrealistic
: parameter, the symbol set is 27 characters (26 english letters and a
: space) whereas in real organisms the symbol set is 4 characters
: (ATGC), which would make the program go _much_ faster (even if you
: used the protein symbol set of 20 characters it would still go
: significantly faster).
I hadn't realized that before. Quite significant. Thanks
for your ongoing efforts to explain these issues.
Jim Acker
*** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** ***
Jim Acker
jac...@gl.umbc.edu
A second flood, a simple famine, plagues of locusts everywhere,
Or a cataclysmic earthquake, I'd accept with some despair.
But no, you sent us Congress! Good God, sir, was that fair?
--- John Adams, "Piddle, Twiddle, and Resolve", from the
musical "1776"
Ian Musgrave & Peta O'Donohue
Address altered to avoid spam, delete RemoveInsert
On Mon, 19 Aug 2002 17:58:48 +0000 (UTC), James Acker
<jac...@linux2.gl.umbc.edu> wrote:
>Ian Musgrave & Peta O'Donohue <reynella_R...@werple.mira.net.au> wrote:
>: While Mr. ReMine changed a perceived "unrealistic" parameter which
>: slowed the program down, he did not change another unrealistic
>: parameter, the symbol set is 27 characters (26 english letters and a
>: space) whereas in real organisms the symbol set is 4 characters
>: (ATGC), which would make the program go _much_ faster (even if you
>: used the protein symbol set of 20 characters it would still go
>: significantly faster).
In Dawkins original description he achieved his target in 48
generations on average.
over 8 runs, using a mutation rate of 0.035, an offspring number of
100 and an alphabet of 26 english letters plus space, the average time
to achieve the target was 71 generations.
over 8 runs, using a mutation rate of 0.035, an offspring number of
100 and an alphabet of 20 amino acid symbols plus space, the average
time to achieve the target was 54 generations.
If neutral substitution (eg, any hydrophobic amino acid symbol could
stand in for any other hydrophobic amino acid symbol) was allowed, the
target would be achieved even faster.
Of course, you could go on about this _ad_nauseum_ pointing out where
the program is different from real natural selection, but that is to
obscure the point that this was a quick and dirty thought experiment
to demonstrate the difference between _cumulative_ selection and
random string generation. Whether the string is reached in 48, 54, 71
or even 7,000 generations, it is still much shorter than the roughly
10^32 generations it would take to generate the same string via random
search.
Trying to prentend that this program is a simulation of natural
selection, rather than an illustrrtive thought experiment, is pure
fantasy.
> I hadn't realized that before. Quite significant. Thanks
>for your ongoing efforts to explain these issues.
My pleasure.
Roy Thearle
[...]
> In other words, Dawkins used a reproduction rate, N, of **AT LEAST** 100!
> (In a sexual species that would require females to average 200 progeny
> each.)
> That's correct -- Dawkins didn't tell the public about the high
> reproduction rate (N >= 100) that he used.
> To begin my research, I had first sent Dawkins a check,
> requesting a copy of his program. He kindly wrote back, and returned the
> check, saying the simulation had been "lost."
Last week you told us that the default size of Dawkin's program was
100.
Now you say that Dawkins didn't publicise it, and has lost the
original program . But you still say that it was 100 or more.
So:
1) How do you know what the default size was if you haven't seen the
program?
2) Why have you changed your claim from "N=100" to "N>=100"?
Roy
Bobby D. Bryant
3) When are you going to answer my two questions?
Bobby Bryant
Austin, Texas
Ian Musgrave & Peta O'Donohue
Address altered to avoid spam, delete RemoveInsert
On Sat, 17 Aug 2002 10:04:37 +0000 (UTC), David Wilson
<see_sig@for_my.address> wrote:
>In article <3muilucli171ojklt...@4ax.com> on August 13th
>in talk.origins "Ian Musgrave & Peta O'Donohue"
><reynella_R...@werple.mira.net.au> wrote:
>
>> YOU [Walter Remine--djw] have claimed,
>> repeatedly, that Dawkins program demonstrates Haldane's dilemma.
>> ...
>
>As you are aware, the program Mr Remine used was actually written by
>David Wise. The one Dawkins wrote may well have differed from it
>(though probably in relatively unimportant ways). By continually
>attributing this program and its default parameters to Dawkins, Mr Remine
>both misprepresents Dawkins and deprives Wise of the proper credit due to
>him.
Yes, sorry about that, in the past I used to be careful to say "David
Wise's implementation of Dawkins' program" or some such construction.
Now days I have been less careful, in part due to trying to get Mr
ReMine to understand that the program cannot do what he claims it
does. Your point is important, and I will be more careful in future.
[snip yet more evidence that "weasel" style programs do not illustrate
Haldane's Dilemma
David Wilson
in talk.origins Walter ReMine <sci...@nospam.net> wrote:
> David Wilson <see_sig@for_my.address> wrote in
> news:200208161436...@funnelwebinternet.com.au:
>
> > ....
> > Nor can I find anything in Dawkins's description
> > to indicate that he used exactly one mutation per progeny, as claimed
> > by Walter ReMine.
> > ....
> > While I am unaware of anything that would specifically contradict Mr
> > Remine's claim that Dawkins used 100 offspring per parent, I have also
> > seen no evidence that would justify it either.
>
> Wilson suggests far more confusion than is warranted by the evidence. ...
I suggested no such thing. I merely pointed out that statements made
by William Dembski and Walter Remine about the details of Dawkins's
cumulative selection procedure did not appear to be supported by
anything Dawkins wrote in "The Blind Watchmaker".
> ... As stated in my book, merely change the simulation's unrealistic
> deterministic mutation (of EXACTLY one mutation per progeny) ...
>
Pardon me for a moment while I pick my jaw up off the floor.
Now let me get this straight: Mr Remine admits openly that he has no
evidence whatever for his assertion that _Dawkins's_ program implemented
exactly one mutation per progeny, and yet he's still prepared to
propagate that assertion as if it were an established fact? Is that
really correct?
> ... [ change the simulation's unrealistic deterministic mutation ...]
> to a more realistic NON-deterministic mutation (while keeping the
> mutation RATE the same as before, and reproduction rate N=100) --
> this nearly DOUBLES the time needed to reach the target phrase.
>
> In other words, Dawkins used a reproduction rate, N, of **AT LEAST** 100!
> ...
Besides being a complete non-sequitur, this conclusion simply does not stand
up to scrutiny. Here are the expected numbers of generations for the
program to reach the target string for offspring numbers of 70, 80 and 90
when the option to change at most one letter per progeny is chosen:
Number of offspring: 70 80 90
Expected #generations
to reach target: 58.48 53.36 49.54
Standard deviation: 14.11 12.10 10.66
Any of these sets of figures is quite consistent with the lengths of the
runs which Dawkins refers to in "The Blind Watchmaker" (viz. 41, 43 and 64).
So even if he _did_ use the procedure implemented by this option of Wise's
program, it would appear that the population size he used could well have
been as low as 70, or perhaps even a little lower (these conclusions should
be considered tentative, however, since a complete likelihood analysis of
_all_ the data which Dawkins gives could well show that some of these
offspring numbers are in fact much less likely than others).
> ... David Wise's program is the evolutionists' own best reconstruction
> of Dawkins's program, and reproduces Dawkins's published results exactly.
> ...
What on earth are you talking about? The processes implemented by both
Dawkins's and Wise's program are stochastic. There is no way you can
tell whether Wise's program "reproduces Dawkins's published results
exactly" by merely examining its output. I have just now run MONKEY
three times using Dawkins's phrase "METHINKS IT IS LIKE A WEASEL" and
the program defaults for all the other options. It reached the target
in 58, 39 and 35 generations. These are nowhere near "exactly"
Dawkin's published results.
While the behaviour of Wise's program with this combination of options
and parameter values is certainly _consistent_ with Dawkins's results
(as I have already pointed out) so is it also when run with a wide range
of other combinations of options and parameter values (as I have also
pointed out already).
> ... So long as Dawkins's simulation retains any luster in the public mind,
> it will remain the target of my criticism. Evolutionists' attempts to
> divert this to a discussion of "David Wise's program" is, in my view,
> an attempt to allow Dawkins's illusions to continue.
It has never been necessary for "evolutionists", or indeed anyone else,
to try to "divert" any criticisms of Mr Remine's to a discussion of Wise's
program. Every time I have ever seen him try to criticise Dawkins's
alleged "simulation", it has always been _precisely_ the behaviour of
this program of Wise's at which his attempted criticisms have been directed.
---------------------------------------------------------------------
David Wilson
SPAMMERS_fingers@SHOULD_BE_funnelweb_PROSECUTED_internet.com.au
Walter ReMine
> While Haldane's argument is inapplicable to the
> procedures implemented in [Dawkins's] program,
I say the cost of substitution has been confused and garbled into oblivion,
and Wilson's comment is an example of that. In order to achieve
beneficial evolution, selection must have something to select from, and
that requires reproductive excess -- a growth from 'few' to 'many' in
number. This requirement is the cost of substitution, and it is
UNAVOIDABLE. A cost of substitution applies to ANY even remotely realistic
simulation of natural selection. Dawkins's simulation demonstrates the
cost of substitution in action. It demonstrates that decreasing the long-
term average reproduction rate slows the long-term average substitution
rate.
> The failure of Haldane's argument _for this type of process_ can be
> illustrated spectacularly, however, by reducing the alphabet size.
> With an alphabet size of 2, ....
By changing to an alphabet size of 2 he has increased the chance -- that a
given mutation is BENEFICIAL -- TO ~50 PERCENT! Wowsers! And this
simultaneously reduced the chance that it is harmful. No wonder the
simulated evolution speeds up!
From the point of view of cost theory he has changed the simulation, as
follows: By lowering the harmful mutation rate, he has lowered the "cost
of mutation" (more precisely called, the cost of harmful mutation),
therefore, LESS of the fixed reproduction rate is taken for paying this
cost. Therefore MORE of the fixed reproduction rate is left over, and
available to pay the cost of substitution -- this speeds up the long-term
average substitution rate.
In this way, even though the reproduction rate and the cost of substitution
are both unchanged, an increase in the substitution rate can be achieved by
LOWERING the cost of mutation (via unrealistic alterations in mutation
rates). Cost theory is confirmed, not violated, by this example.
> .... By continually attributing this program and its
> default parameters to Dawkins, Mr ReMine ...
> misprepresents Dawkins ...
Wilson has shown no such misrepresentation.
Walter ReMine
news:pln1mu04cntt7ss30...@4ax.com:
> While Mr. ReMine changed a perceived "unrealistic" parameter which
> slowed the program down, he did not change another unrealistic
> parameter, the symbol set is 27 characters (26 english letters and a
> space) whereas in real organisms the symbol set is 4 characters
> (ATGC), which would make the program go _much_ faster (even if you
> used the protein symbol set of 20 characters it would still go
> significantly faster).
Musgrave replaces one of Dawkins's "unrealistic" parameters with a
supposedly less "unrealistic" parameter that makes the simulation run
faster. We are thereby led to believe that Dawkins's simulation did not
favor evolution after all. That is one fat illusion.
Under the initial conditions setup by Dawkins, a given mutation has one
chance in 27 of being beneficial! Wowsers!!! (And with an incredibly
high selection coefficient s=1!) Both those figures are astoundingly
higher (and more favorable to evolution) than in natural populations.
Indeed this OVERCOMES the disadvantages a small population would
ordinarily face. Dawkins's simulation has the evolutionary advantages
of small and large populations AT THE SAME TIME! No natural population
can do that. This simulation is wildly unrealistic IN FAVOR OF
EVOLUTION. Yet it produces a speed less than five times faster than
Haldane's speed limit.
Musgrave says "in real organisms the symbol set is 4 characters (ATGC)"
so, *when used in the program,* the corresponding chance of being
beneficial is one in four -- a 25 percent chance of being beneficial!
He is grotesquely mistaken -- that is not remotely more realistic. In
effect, he proposes a simulation EVEN MORE WILDLY UNREALISTIC IN FAVOR
OF EVOLUTION, then he says this "would make the program go _much_
faster". Once you understand his mistake, its faster speed further
supports my point.
Musgrave has identified several things about Dawkins's simulation that
are "unrealistic" -- but they ALL unrealistically FAVOR evolution and
its speed.
>>Wilson suggests far more confusion than is warranted by the evidence.
>>As stated in my book, merely change the simulation's unrealistic
>>deterministic mutation (of EXACTLY one mutation per progeny) to a more
>>realistic NON- deterministic mutation (while keeping the mutation RATE
>>the same as before, and reproduction rate N=100) -- this nearly
>>DOUBLES the time needed to reach the target phrase.
>
> However, the whole point is that there is no evidence that Dawkins
> _did_ use exactly one mutation per progeny.
There is profound evidence that Dawkins's simulation is unrealistically
skewed (in over a dozen ways) to favor evolution, and that he did NOT
reveal them to the public.
I did reveal them to the public.
> Dawkins program was simply contrasting a toy mutation and selection
> system with a "typing monkey" random string generator. The point being
> that mutation and selection was much, much faster by many orders of
> magnitude ....
Dawkins and evolutionists made SPEED an issue, therefore I am justified
in dismantling that illusion.
> No, [Dawkins] drew attention to the _main_ reason that
> the program did not mimic natural selection in the wild
> (and pointed out that it was not the only reason either).
Dawkins's "main reason" is only a token, compared to the many (over a
dozen) other reasons why his simulation is unrealistic in favor of
evolution. He silently assumed-away anything that could prevent
"success" of reaching the target, and he further silently assumed many
things that increased its speed. His simulation was ripe for debunking,
and I did so.
Walter ReMine
wrote in news:k9i3mukjd1v4d0ufc...@4ax.com:
> Of course, you could go on about this _ad_nauseum_ pointing out where
> the program is different from real natural selection, but that is to
> obscure the point that this was a quick and dirty thought experiment
> ....
> Trying to pretend that this program is a simulation of natural
> selection, rather than an illustrative thought experiment, is pure
> fantasy.
Ian Musgrave has attacked me viciously (and his webpage STILL attacks me)
on the grounds that (in Dawkins's simulation) "SIZE MATTERS". I keep
agreeing, "size matters" and it's IN FAVOR OF EVOLUTION!
He cannot honestly attack me on such grounds.
His going on "ad nauseum" about unrealistic features of the simulation
"OBSCURES THE POINT" that they are all in FAVOR OF EVOLUTION.
Dawkins's simulation is wildly unrealistic in FAVOR of evolution. Yet it
produces a speed less than five times faster than Haldane's speed limit.
Calling it "not a simulation" does not let evolutionists off the hook.
Walter ReMine
WEASEL program -- and both are grounded in his continued misrepresentations
of my book (by gross omission):
1) -- Musgrave notes several unrealistic things about the simulation, but
has not shown that they DIS-favor evolution. He says Dawkins's program is
an *unrealistic* simulation, and "SIZE MATTERS". I keep agreeing -- size
matters, and its in FAVOR of evolution in this simulation.
Dawkins's simulation is wildly unrealistic (in over a dozen ways) -- IN
FAVOR OF EVOLUTION! Yet the simulation reaches a speed less than five
times faster than Haldane's Dilemma. Simply calling it "not a simulation"
does not let evolutionists off the hook.
Yes, "size matters" but Dawkins's simulation is neatly designed to disallow
the evolutionary disadvantages that small populations would ordinarily
face. Instead, the simulation has the evolutionary advantages of large AND
small populations AT THE SAME TIME! No natural population can do that!
Musgrave pretends to engage my argument, when he actually omits it
entirely.
2) -- In this post I discuss (as in 1999) his second type of criticism. He
says (contrary to me) that Dawkins's WEASEL program cannot demonstrate
ANYTHING about Haldane's Dilemma. His reasoning is based entirely on his
false definition of Haldane's Dilemma.
Here is my brief definition of Haldane's Dilemma: Cost theory predicts
that a reduction in long-term average reproduction rate will reduce the
long-term average substitution rate. The long-term average cost of
substitution (which Haldane estimated at 30), together with the species'
long-term average reproductive excess (which Haldane estimated at 0.1, as
acknowledged by various evolutionary commentators) would limit the
beneficial substitution rate to one per 300 generations. That rate means
that in ten million years a human-ape-like population could achieve no more
than 1667 substitutions (where each is nominally one nucleotide, under
current evolutionary views). This would be the maximum allowed for
explaining the origin of ALL beneficial human changes arising over that ten
million years: upright posture, the tripling of brain size, speech,
language, and more.
Here is Musgrave's definition:
> Haldane's "dilemma" is that, with certain assumptions, and in certain
> broad ranges, it will take just as long to "fix" (ie go from being
> very rare to being present in all members of the population) two
> simultaneous beneficial mutations (ie the organism have two benefical
> mutations in the same organism) in a population as it takes to fix two
> sequential mutations (ie that it takes longer to fix a beneficial
> double mutant than a beneficial single mutant see R Williams page in
> the links section of my page for mathematical discussion of why this
> should be so).
1) His definition is unrevealing to the public. It is downright obscure.
An ordinary person would get little idea of what the problem REALLY is.
Such needlessly obscure discussions are why my expose' on Haldane's Dilemma
is needed.
2) His definition is peculiar and atypical even by the standards of the
traditional literature.
3) Most importantly, his definition is overly narrow (and thereby mistaken)
-- it reflects only a tiny facet of the problem, and not the core of it.
The dilemma, in Musgrave's view, is that "it will take just as long to fix
two simultaneous beneficial mutations in a population as it takes to fix
two sequential mutations." Notice his emphasis on: (a) the time it takes
to complete a SINGLE substitution (which omits the time BETWEEN
substitutions), and (b) SEQUENTIAL versus SIMULTANEOUS substitutions.
His faulty definition of Haldane's Dilemma leads him to the following
interpretation.
> Mr. ReMine claims that Dawkins weasel program demonstrates Haldane's
> dilemma. Yet in Dawkins program (and the majority of those programs
> that replicate Dawkins original), beneficial mutations are
> automatically fixed in a _constant_ 1 generation. Dawkins program (and
> Dawkins-style programs, such as on my page) CANNOT demonstrate
> anything about Haldane's dilemma.
And:
> .... Importantly, Dawkins
> program CANNOT show Haldane's dilemma, as the rate of fixation of
> beneficial mutations is _always_ 1 generation, regardless of whether
> there are one, two or more beneficial mutations present in a single
> organism (recall that Haldane's dilemma says it will take longer to
> fix two simultaneous beneficial mutations than to fix one beneficial
> mutation).
In Dawkins's program a beneficial mutation substitutes in EXACTLY 1
generation -- therefore, according to Musgrave, "Dawkins' program CANNOT
demonstrate anything about Haldane's Dilemma." Notice he ignores the LONG-
TERM AVERAGE substitution rate, which is what Haldane's Dilemma is about.
Also notice he ignores the relationship between reproduction rate and long-
term average substitution rate. Dawkins's program demonstrates this
phenomenon, and error catastrophe, and a speed comparable to Haldane's
Dilemma -- all of which helps my readers understand the issues.
-- Walter ReMine
James Acker
[deletions]
: Yes, "size matters" but Dawkins's simulation is neatly designed to disallow
: the evolutionary disadvantages that small populations would ordinarily
: face. Instead, the simulation has the evolutionary advantages of large AND
: small populations AT THE SAME TIME! No natural population can do that!
: Musgrave pretends to engage my argument, when he actually omits it
: entirely.
: 2) -- In this post I discuss (as in 1999) his second type of criticism. He
: says (contrary to me) that Dawkins's WEASEL program cannot demonstrate
: ANYTHING about Haldane's Dilemma. His reasoning is based entirely on his
: false definition of Haldane's Dilemma.
: Here is my brief definition of Haldane's Dilemma: Cost theory predicts
: that a reduction in long-term average reproduction rate will reduce the
: long-term average substitution rate. The long-term average cost of
: substitution (which Haldane estimated at 30), together with the species'
: long-term average reproductive excess (which Haldane estimated at 0.1, as
: acknowledged by various evolutionary commentators) would limit the
: beneficial substitution rate to one per 300 generations. That rate means
: that in ten million years a human-ape-like population could achieve no more
: than 1667 substitutions (where each is nominally one nucleotide, under
: current evolutionary views). This would be the maximum allowed for
: explaining the origin of ALL beneficial human changes arising over that ten
: million years: upright posture, the tripling of brain size, speech,
: language, and more.
Evolutionary biologist Scott Page writes:
"According to ReMine's application of Haldane's model, no more than
1667 such mutations could have become fixed in 10 million years. As
Haldane admitted in his 1957 paper, and alluded to in his 1960 paper,
his numbers would probably require 'drastic revision', and, as he pointed
out, one should not apply his model where it is not applicable.
Among the constraints in Haldane's model were weak selection and a
constant population size. Many evolutionary biologists (e.g., Felsenstein;
Darlington) _demonstrated_ that in many instances, the events that would lead
to a speciation event would induce strong selection (as in some peppered
moth populations) and a reduced population size, such that the cost
would be 'paid' rapidly and the beneficial allele(s) fixed in short order.
In addition, there are documented examples of organisms evolving at rates
which exceed the limit imposed by Haldane's model (e.g., "Population
structure in relation to cost of selection," Grant and Flake,
1974, PNAS 71(5)1670-71). Such counter-examples are mysteriously missing
from ReMine's book."
Y'know, he's right. I've got the book right here, and despite the
numerous references, the Grant and Flake paper isn't listed. But
I'll provide the abstract of the paper above right here:
Abstract: The opposing requirements in an evolving population for a rapid
rate of multiple gene substitution and for the maintenance of
normal population size can be reconciled in a variety of ways. The ways
out of the impasse suggested here invoke deviations
from the usual assumption of a large continuous population with constant
numbers. In a colonial population system there may
be significant random fluctuations in the accidental mortality rate
between different colonies; and those colonies with reduced
numbers of accidental deaths could tolerate the larger number of selective
deaths that go hand in hand with rapid evolution. In
new daughter colonies founded by one or a few colonizing individuals
from a large polymorphic ancestral population, some
genes may reach complete fixation in one or a few generations, without the
usual concomitant selective cost. Or, in the same setup, the favored alleles
may change by chance from rare to moderately common, but not to complete
fixation, during the founding of some daughter colonies; and this raises the
allele frequencies above the low range, where the cost of selection is
greatest, so that the cost of further selective changes is bearable.
Also not listed are these two papers, also by authors
Verne Grant and Robert Flake:
"An Analysis of the Cost of Selection Concept", PNAS, Vol. 71, No. 9. (Sep.,
1974), pp. 3716-3720.
Abstract: It is shown for a continuous haploid model that the common
standard assumptions used in calculating the cost of gene
substitution, namely, large constant population size and small constant
selective value, are unnecessary. Population size may
fluctuate during the course of substitution without affecting the
calculated total cost. The selective intensity does not need to
be small and constant to give the standard result for substitution cost.
Diploid models with multiple alleles are analyzed and
contrasted with standard two-allele models in respect to calculation of
substitution cost. The influence of population structure
on the probability of occurrence of complete gene substitution is
discussed on the basis of a numerical example. The robust
nature of the cost-of-selection concept is examined in the light of a
conservation principle.
"Solutions to the Cost-of-Selection Dilemma" PNAS, Vol. 71, No. 10. (Oct.,
1974), pp. 3863-3865.
Abstract: There are various biologically realistic ways around the
cost-of-selection restriction on rapid multiple-gene substitution. Some
of these ways depend upon particular forms of interaction or linkage in the
genes undergoing substitution; other ways depend
on particular conditions of population size and structure. The special
genotypic and populational conditions required for rapid
evolutionary change in genetically complex characters are not unusual in
higher organisms.
[That last sentence sure is a kick, isn't it?]
One would think that a scholarly treatment of Haldane's Dilemma
predicated on "cost theory" would at least mention these papers.
Particularly when the author makes this claim about Haldane's
Dilemma:
"evolutionists obscured it into oblivion and prematurely brushed it
aside -- ESPECIALLY when addressing the generally public."
I think the one doing the obscuration of what the scientists
have actually been doing is the author of "The Biotic Message".
Jim Acker
*** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** ***
Jim Acker
jac...@gl.umbc.edu
"Since we are assured that an all-wise Creator has observed the
most exact proportions, of number, weight, and measure, in the
make of all things, the most likely way therefore, to get any
insight into the nature of those parts of the creation, which
come within our observation, must in all reason be to number,
weigh, and measure." - Stephen Hales
MEC
Mr Remine
As a lurker I have witnessed this back and forth for....what....three
years now? Frankly I am flabbergasted that you STILL don't get it. As
Ian and others have been trying to tell you -and which everyone else
with passing knowledge of this thread knows except apparently
(lamentably) you- Dawkin's program CANNOT, DID NOT, NEVER WAS INTENDED
TO, IS INCAPABLE OF modeling how evolution *actually* works. He
(Dawkins) and subsequently David Wise, was simply trying to illustrate
the power of selection over random searches. No more, no less.
Ian and others are also correct in saying that Dawkin's simulation
CANNOT, DID NOT, NEVER WAS INTENDED TO, IS INCAPABLE OF making
insights into Haldane's dilemma.
Give it up, Mr. Remine. Your diversions and misreadings as illustrated
above do you no justice. If you want to gain (or keep) the interest of
lurkers it would pay for you to answer Bobby Bryants' two questions.
Here they are;
/
a) What point was Dawkins trying to make when he described WEASEL?
b) Is WEASEL a good way to make that point?
//
Here's the post (watch the wrap!)
James Acker
Follow-up:
James Acker <jac...@linux2.gl.umbc.edu> wrote:
"Population
: structure in relation to cost of selection," Grant and Flake,
: 1974, PNAS 71(5)1670-71).
: "An Analysis of the Cost of Selection Concept", PNAS, Vol. 71, No. 9. (Sep.,
: 1974), pp. 3716-3720.
: "Solutions to the Cost-of-Selection Dilemma" PNAS, Vol. 71, No. 10. (Oct.,
: 1974), pp. 3863-3865.
After my post, I realized that it may be difficult for Mr. ReMine to
obtain copies of these papers. It is not difficult for me to obtain copies
of them. As a matter of fact, I now have them in hand.
Therefore, if Mr. ReMine makes the request, I will be glad
to fully undertake the costs of sending Mr. ReMine copies of these
three papers in order to assist his research on Haldane's Dilemma,
which is funded by the Discovery Institute of Science and Culture
(they changed their name and took out "Renewal").
Unless otherwise instructed I would send these papers to the
St. Paul Science Inc. address listed on the "Biotic Message" Web
site.
Jim Acker
*** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** ***
Jim Acker
jac...@gl.umbc.edu
Ian Musgrave & Peta O'Donohue
Address altered to avoid spam, delete RemoveInsert
On Thu, 22 Aug 2002 05:00:15 +0000 (UTC), Walter ReMine
<sci...@nospam.net> wrote:
>Ian Musgrave poses two types of criticism against my material on Dawkins's
>WEASEL program -- and both are grounded in his continued misrepresentations
>of my book (by gross omission):
>
>1) -- Musgrave notes several unrealistic things about the simulation, but
>has not shown that they DIS-favor evolution. He says Dawkins's program is
>an *unrealistic* simulation, and "SIZE MATTERS". I keep agreeing -- size
>matters, and its in FAVOR of evolution in this simulation.
Your claim is that the program demonstrates Haldane's Dilemma. It
doesn't, it can't. You can make the program go slow, but _why_ it goes
slow is irrelevant to Haldane's dilemma.
>Dawkins's simulation is wildly unrealistic (in over a dozen ways) -- IN
>FAVOR OF EVOLUTION! Yet the simulation reaches a speed less than five
>times faster than Haldane's Dilemma. Simply calling it "not a simulation"
>does not let evolutionists off the hook.
>
>Yes, "size matters" but Dawkins's simulation is neatly designed to disallow
>the evolutionary disadvantages that small populations would ordinarily
>face. Instead, the simulation has the evolutionary advantages of large AND
>small populations AT THE SAME TIME! No natural population can do that!
>Musgrave pretends to engage my argument, when he actually omits it
>entirely.
No, I focus on the key issue which you still don't understand. In
Wise's (and others) implementation of the weasel program beneficial
mutations are fixed in a constant one generation, regardless of the
offspring size (the breeding population is exactly one sting at all
times). It doesn't benefit in any of the ways Mr. ReMine thinks it
does.
>2) -- In this post I discuss (as in 1999) his second type of criticism. He
>says (contrary to me) that Dawkins's WEASEL program cannot demonstrate
>ANYTHING about Haldane's Dilemma. His reasoning is based entirely on his
>false definition of Haldane's Dilemma.
Please read Haldane's original paper. My usage is Haldane's usage
(except he gives an illustration with 10 beneficial mutations rather
than two, as his lead in example). It is also instructive to read in
his conclusions where he points out that his average rate is
compatible with the average rate of evolution seen in the fossil
record.
>Here is my brief definition of Haldane's Dilemma: Cost theory predicts
>that a reduction in long-term average reproduction rate will reduce the
>long-term average substitution rate. The long-term average cost of
>substitution (which Haldane estimated at 30), together with the species'
>long-term average reproductive excess (which Haldane estimated at 0.1, as
>acknowledged by various evolutionary commentators) would limit the
>beneficial substitution rate to one per 300 generations.
And they are _estimates_. Haldane's "speed limit" is not a fixed
limit. It's an average rate under a given set of assumptions (hard
selection, lowish selection intensity, independent mutations, no
genetic bottle necks). These assumptions are often violated in the
real world, and it's not hard to find real world examples where
natural selection produces rates of genetic change faster than
Haldane's limit (See the list of references in James Ackers excellent
post).
Haldane's dilemma is NOT the number 300 generations (which he and
others felt was consistent with known rates of evolution in the fossil
record), but that n simultaneous beneficial mutations would take as
long to fix in a population as n sequential mutations. His
illustration with the peppered moth on the first page is quite clear
about this.
Haldane also pointed out that he expected research to find exceptions
to his assumptions, and it has. As just one example, gene
"hitchhiking" is an often observed real world example where multiple
genes go to fixation by "hitchhiking" with a beneficial gene (see also
the literature on selective sweeps).
>That rate means
>that in ten million years a human-ape-like population could achieve no more
>than 1667 substitutions (where each is nominally one nucleotide, under
>current evolutionary views). This would be the maximum allowed for
>explaining the origin of ALL beneficial human changes arising over that ten
>million years: upright posture, the tripling of brain size, speech,
>language, and more.
As it looks like only 50 gene changes are needed to explain our brain
size, language etc. this is more than enough. A single gene controls
brain size, neuronal connectivity and formation of infolding. A single
gene FOX-32, seems to be critical for the precise motor function
needed for advanced speech. There are but two nucleotide substitutions
between us and apes in FOX-32.
1667 beneficial gene substitutions are enough, probably more than
enough.
>Here is Musgrave's definition:
>
>> Haldane's "dilemma" is that, with certain assumptions, and in certain
>> broad ranges, it will take just as long to "fix" (ie go from being
>> very rare to being present in all members of the population) two
>> simultaneous beneficial mutations (ie the organism have two benefical
>> mutations in the same organism) in a population as it takes to fix two
>> sequential mutations (ie that it takes longer to fix a beneficial
>> double mutant than a beneficial single mutant see R Williams page in
>> the links section of my page for mathematical discussion of why this
>> should be so).
>
>1) His definition is unrevealing to the public. It is downright obscure.
What is obscure about it?
>An ordinary person would get little idea of what the problem REALLY is.
That _is_ the problem, the core of Haldane's dilemma. Not the figure
300 generations, which was generated with a particular set of
assumptions that is often violated in the real world. it is WHY the
figure 300 generations turns up when you use that set of assumptions.
>Such needlessly obscure discussions are why my expose' on Haldane's Dilemma
>is needed.
It is _your_ expose that is needlessly obscure.
>2) His definition is peculiar and atypical even by the standards of the
>traditional literature.
It's how Haldane originally presented it, and the key issue in most of
the treatments of the dilemma. For example, gene hitchhiking and
truncation selection directly address this issue.
>3) Most importantly, his definition is overly narrow (and thereby mistaken)
>-- it reflects only a tiny facet of the problem, and not the core of it.
>The dilemma, in Musgrave's view, is that "it will take just as long to fix
>two simultaneous beneficial mutations in a population as it takes to fix
>two sequential mutations." Notice his emphasis on: (a) the time it takes
>to complete a SINGLE substitution (which omits the time BETWEEN
>substitutions), and (b) SEQUENTIAL versus SIMULTANEOUS substitutions.
Yes, this was Haldane's point. If you could fix two or more
beneficial mutations simultaneously at the same rate as a single
muation was fixed, then evolution could run really fast. See his
peppered moth example.
>His faulty definition of Haldane's Dilemma leads him to the following
>interpretation.
>
>> Mr. ReMine claims that Dawkins weasel program demonstrates Haldane's
>> dilemma. Yet in Dawkins program (and the majority of those programs
>> that replicate Dawkins original), beneficial mutations are
>> automatically fixed in a _constant_ 1 generation. Dawkins program (and
>> Dawkins-style programs, such as on my page) CANNOT demonstrate
>> anything about Haldane's dilemma.
Yep.
>And:
>
>> .... Importantly, Dawkins
>> program CANNOT show Haldane's dilemma, as the rate of fixation of
>> beneficial mutations is _always_ 1 generation, regardless of whether
>> there are one, two or more beneficial mutations present in a single
>> organism (recall that Haldane's dilemma says it will take longer to
>> fix two simultaneous beneficial mutations than to fix one beneficial
>> mutation).
Yep.
>In Dawkins's program a beneficial mutation substitutes in EXACTLY 1
>generation -- therefore, according to Musgrave, "Dawkins' program CANNOT
>demonstrate anything about Haldane's Dilemma." Notice he ignores the LONG-
>TERM AVERAGE substitution rate, which is what Haldane's Dilemma is about.
No, what you are seeing is the long term rate of _appearance_ of
beneficial mutations NOT the substitution rate. The substitution rate
is ALWAYS one. When a mutation appears, it is immediately fixed. Late
in the program, it takes a while for a beneficial mutation to turn up,
but it is still fixed in one generation. That you confuse the mutation
appearance rate with substitution rate speaks volumes about your
understanding of the biology involved.
>Also notice he ignores the relationship between reproduction rate and long-
>term average substitution rate. Dawkins's program demonstrates this
>phenomenon, and error catastrophe, and a speed comparable to Haldane's
>Dilemma -- all of which helps my readers understand the issues.
No. All you have demonstrated is that in a population with only one
breeding "organism", when that organism has only 6 offspring, it takes
a long time for a beneficial mutation to show up. This is irrelevant
to Haldane's Dilemma, and shows your ignorance of the program and the
Dilemma.
Cheers! Ian
Ian Musgrave & Peta O'Donohue
Address altered to avoid spam, delete RemoveInsert
On Thu, 22 Aug 2002 04:50:12 +0000 (UTC), Walter ReMine
<sci...@nospam.net> wrote:
>"Ian Musgrave & Peta O'Donohue" <ian.musgr...@adelaide.edu.au>
>wrote in news:k9i3mukjd1v4d0ufc...@4ax.com:
>
>> Of course, you could go on about this _ad_nauseum_ pointing out where
>> the program is different from real natural selection, but that is to
>> obscure the point that this was a quick and dirty thought experiment
>> ....
>> Trying to pretend that this program is a simulation of natural
>> selection, rather than an illustrative thought experiment, is pure
>> fantasy.
>
>Ian Musgrave has attacked me viciously (and his webpage STILL attacks me)
>on the grounds that (in Dawkins's simulation) "SIZE MATTERS". I keep
>agreeing, "size matters" and it's IN FAVOR OF EVOLUTION!
No, if I were to attack you, I would say that you were an otiose
purveyor of pernicious persiflage lower than John Howard's eyebrows.
However, I haven't attacked you. I have pointed out your lack of
understanding of both implementations of Dawkins program and Haldane's
dilemma. Far worse occurs during peer review of publications and
grants. Scientists in general have to face severe criticism of their
ideas, Mr. ReMine should face up to the criticism of his claims and
actually deal with this issues raised.
>He cannot honestly attack me on such grounds.
I honestly point out that you don't understand what is going on. In
Wise's (and my, and others) implementation of the Weasel program,
beneficial mutations are always fixed in a constant one generation. It
doesn't matter whether there are one two or more simultaneous
beneficial mutations, the mutations are always fixed in a single
generation. This CANNOT demonstrate Haldane's dilemma, as the core of
the dilemma is that it takes as long to fix two simultaneous
beneficial mutations as it does to fix two sequential mutations.
>His going on "ad nauseum" about unrealistic features of the simulation
>"OBSCURES THE POINT" that they are all in FAVOR OF EVOLUTION.
Ahem, I gave an example where the program was not explicitly NOT in
favor of evolution.
More to the point, you can fiddle with the program and alter speeds
till the cows come home. It still doesn't demonstrate Haldane's
dilemma.
Haldane's dilemma is about the time it takes to fix one or more
beneficial mutations. In the programs under discussion, beneficial
mutations are fixed in a constant one generation. How fast
implementations of the Weasel program go are irrelevant.
>Dawkins's simulation is wildly unrealistic in FAVOR of evolution. Yet it
>produces a speed less than five times faster than Haldane's speed limit.
Haldane's "speed limit" is not a fixed limit. It's an average rate
under a given set of assumptions (hard selection, lowish selection
intensity, independent mutations, no genetic bottle necks). These
assumptions are often violated in the real world, and it's not hard to
find real world examples where natural selection produces rates of
genetic change faster than Haldane's limit (See the list of references
in James Ackers excellent post).
>Calling it "not a simulation" does not let evolutionists off the hook.
It's not a simulation of natural selection, but an illustration of the
difference between random string generation and cumulative selection.
That Mr. ReMine thinks that the speeds of change in this program are
illustrative of Haldanes's dilemma is a sad indictment of his
understanding of the program and evolutionary biology.
Ian Musgrave & Peta O'Donohue
Address altered to avoid spam, delete RemoveInsert
On Thu, 22 Aug 2002 04:29:04 +0000 (UTC), Walter ReMine
<sci...@nospam.net> wrote:
>David Wilson <see_sig@for_my.address> wrote in
>news:200208171027...@funnelwebinternet.com.au:
>
>
>> While Haldane's argument is inapplicable to the
>> procedures implemented in [Dawkins's] program,
>
>I say the cost of substitution has been confused and garbled into oblivion,
>and Wilson's comment is an example of that.
No, it hasn't. It is discussed in the technical literature in
technical terms (as is standard selection, neutral drift etc.). That
you don't understand it is immaterial.
However, David Wilsons comment directly addresses a key issue in clear
and non-technical terms.
>In order to achieve
>beneficial evolution, selection must have something to select from, and
>that requires reproductive excess -- a growth from 'few' to 'many' in
>number. This requirement is the cost of substitution, and it is
>UNAVOIDABLE.
But not a constant, nor necessarily very large, in the real world.
>A cost of substitution applies to ANY even remotely realistic
>simulation of natural selection. Dawkins's simulation demonstrates the
>cost of substitution in action. It demonstrates that decreasing the long-
>term average reproduction rate slows the long-term average substitution
>rate.
No, it doesn't. Cost of substitution is immaterial in Wise's (and
others) implementation of Dawkins program. Beneficial mutations are
fixed in a constant one generation.
>> The failure of Haldane's argument _for this type of process_ can be
>> illustrated spectacularly, however, by reducing the alphabet size.
>> With an alphabet size of 2, ....
>
>By changing to an alphabet size of 2 he has increased the chance -- that a
>given mutation is BENEFICIAL -- TO ~50 PERCENT! Wowsers! And this
>simultaneously reduced the chance that it is harmful. No wonder the
>simulated evolution speeds up!
You have missed the point. In the part you have snipped, Mr Wilson
demonstrated how you can falsify a prediction from Haldane's
formulation. The issue is not the speed at which the process occurs,
but the fact that Haldane's rules give the wrong answer in the
situation described.
>From the point of view of cost theory he has changed the simulation, as
>follows: By lowering the harmful mutation rate, he has lowered the "cost
>of mutation" (more precisely called, the cost of harmful mutation),
>therefore, LESS of the fixed reproduction rate is taken for paying this
>cost. Therefore MORE of the fixed reproduction rate is left over, and
>available to pay the cost of substitution -- this speeds up the long-term
>average substitution rate.
No, the substitution rate is always a constant one generation.
>In this way, even though the reproduction rate and the cost of substitution
>are both unchanged, an increase in the substitution rate can be achieved by
>LOWERING the cost of mutation (via unrealistic alterations in mutation
>rates). Cost theory is confirmed, not violated, by this example.
Go back and run through the example again, carefully.
>> .... By continually attributing this program and its
>> default parameters to Dawkins, Mr ReMine ...
>> misprepresents Dawkins ...
>
>Wilson has shown no such misrepresentation.
It is there in black and white. You make claims about Dawkins proram
for which you have no evidence.
Cheers! Ian
Dirk Murcray
> WEASEL program -- and both are grounded in his continued misrepresentations
> of my book (by gross omission):
>
> 1) -- Musgrave notes several unrealistic things about the simulation, but
> has not shown that they DIS-favor evolution. He says Dawkins's program is
> an *unrealistic* simulation, and "SIZE MATTERS". I keep agreeing -- size
> matters, and its in FAVOR of evolution in this simulation.
>
> Dawkins's simulation is wildly unrealistic (in over a dozen ways) -- IN
> FAVOR OF EVOLUTION! Yet the simulation reaches a speed less than five
> times faster than Haldane's Dilemma. Simply calling it "not a simulation"
> does not let evolutionists off the hook.
>
> Yes, "size matters" but Dawkins's simulation is neatly designed to disallow
> the evolutionary disadvantages that small populations would ordinarily
> face. Instead, the simulation has the evolutionary advantages of large AND
> small populations AT THE SAME TIME! No natural population can do that!
> Musgrave pretends to engage my argument, when he actually omits it
> entirely.
Why is it that you cannot grasp the simple point that Dawkins' WEASEL
program was never proposed as a computer simulation of natural
selection? It has been pointed out repeatedly that it was merely a
demonstration of the exponentially increased rates of probability of
cumulative selection compared to repetitive random chance (as opposed
to the impossibly low probabilities of evolution typically argued by
creationists).
Adam Marczyk
news:ak310n$e53$1...@news.umbc.edu...
Since the archive really doesn't have as much material on Haldane's Dilemma as
it should, I nominate this excellent post for POTM. Seconds?
[snip]
--
And I want to conquer the world,
give all the idiots a brand new religion,
put an end to poverty, uncleanliness and toil,
promote equality in all of my decisions...
--Bad Religion, "I Want to Conquer the World"
http://www.ebonmusings.org ICQ: 8777843 PGP Key ID: 0x5C66F737
Adam Marczyk
[snip]
> > Dawkins program was simply contrasting a toy mutation and selection
> > system with a "typing monkey" random string generator. The point being
> > that mutation and selection was much, much faster by many orders of
> > magnitude ....
>
> Dawkins and evolutionists made SPEED an issue, therefore I am justified
> in dismantling that illusion.
Amazingly, you quoted the paragraph that cuts right to the heart of your
continuing fundamental misunderstanding of what the WEASEL program was intended
to do - without giving any sign of comprehending that your misunderstanding even
exists. I can no longer conclude that your obstinate refusal to grasp this point
is anything other than deliberate. The WEASEL program was not *ever* intended to
be a simulation of evolution. Never. Dawkins explicitly says so. Why is it so
hard for you to understand this very simple point?
Harlequin
[snip]
> Since the archive really doesn't have as much material on Haldane's
> Dilemma as it should, I nominate this excellent post for POTM.
> Seconds?
>
> [snip]
>
I will second it.
--
Anti-spam: replace "usenet" with "harlequin2"
Walter ReMine
> .... The WEASEL program was not *ever* intended to be a
> simulation of evolution. Never.
Dawkins's WEASEL program was used to demonstrate natural selection's: (1)
"success", and (2) speed.
Both are illusions my book dismantles.
Walter ReMine
news:c240c53.02082...@posting.google.com:
> .... As Ian [Musgrave] and others have been trying
> to tell you ...
Musgrave said "SIZE MATTERS". He is wrong to leave the argument there. In
Dawkins's simulation, "size matters" IN FAVOR OF EVOLUTION! Musgrave has
no grounds for assaulting me on this. He is repeatedly misrepresenting my
argument (by gross omission).
Musgrave's claims about "Haldane's Dilemma" are false, because they arise
from his false definition of Haldane's Dilemma. Again he is
misrepresenting my argument (by gross omission).
> a) What point was Dawkins trying to make when
> he described WEASEL?
I cannot read Dawkins's mind. But what he DID DO was use his WEASEL
simulation to demonstrate natural selection's: (1) "success" and (2) speed.
BOTH of those are illusions dismantled in my book.
> b) Is WEASEL a good way to make that point?
Dawkins's WEASEL program/discussion is a finely-tuned illusion. It was a
"good way" to mislead the public. My book dismantles it.
Harlequin
> "Adam Marczyk" <ebon...@hotmailNOTexcite.com> wrote in
> news:umbasm1...@corp.supernews.com:
>
>> .... The WEASEL program was not *ever* intended to be a
>> simulation of evolution. Never.
>
> Dawkins's WEASEL program was used to demonstrate natural selection's:
> (1) "success", and (2) speed.
>
> Both are illusions my book dismantles.
Your copy of _The Blind Watchmaker_ must be an entirely different
than mine since Dawkins is quite clear in my copy about the purpose
of the WEASEL program. TBW does not in any way, shape, or
form say that that program showed that natural selection is
a "success" nor does it address the speed of natural selection.
The purpose of the WEASEL program is to explain (not prove) the
difference between single-step and cumulative selection. If
the program is to be said to "demonstrate" anything, it
"demonstrates" that single-step natural selection does not work.
Dawkins clearly states on page 30 that the WEASEL program "is useful
for explaining the distinction between single-step selection
and cumulative selection" though "it is misleading in
important ways." He then gives some reasons why the example
is misleading which leads on to a more sophisticated model, one
which Dawkins used far more pages for, to further help explain
(not prove) how natural selection works.
So what is next, debunking water in pipe analogies in discussions
about electricity since they don't "prove" anything either?
Or how about debunking spots on balloons that the astronomers
use to help explain why distant galaxies are going away from
us faster than galaxies that are closer to us? Putting
spots on a balloon does not prove anything about the universe.
But is it certainly a very useful aide in communicating a concept.
The programs that Dawkins uses in TBW are used for the same purpose.
They don't prove anything, but they certainly are useful
explaining/teaching who natural selection works.
Walter ReMine
> ... I've got [a photocopy of _The Biotic Message_]
> right here, and despite the numerous references,
> the Grant and Flake paper isn't listed.
_The Biotic Message_ does not intend to dissect all the PAPERS on Haldane's
Dilemma. That would be a twisty-turny journey most of my readers would
find boring. Especially when such detours end up at the same destination
as my material. That is, my book covers all the CONCEPTS AND ARGUMENTS on
Haldane's Dilemma (one way or another), and de-emphasizes the needlessly
tedious issue of 'who said what when.' My three chapters aim at (and
succeed in) making Haldane's Dilemma issues understandable to a broad
general audience.
All the fundamental mistakes are illuminated in my book, and there is
little point (other than perhaps an overly perverse academic interest) in
documenting all the *permutations* of ways the fundamental mistakes are
made in the literature. The conceptual mistakes made by "Grant and Flake"
(circa 1974) are not unique to them, and are covered in my book (though not
under their names). (My chapter cites Grant's **1985 book**, yes that's
the same person.)
Indeed, it is more effective when my book cites MORE RECENT and more widely
KNOWN authorities (such as David Merrell, and Ernst Mayr) who survey
Haldane's Dilemma better than Grant and Flake did. I regard David
Merrell's book, _Ecological Genetics_, as the best (and least confused)
survey of Haldane's Dilemma, outside of my book -- and my book cites/quotes
it many times.
I claim Haldane's Dilemma was never solved. It was merely confused,
obscured, and prematurely brushed aside. When evolutionists wave off-hand
at 1974 papers -- rather than offer widely-accepted SOLUTIONS -- that is a
diversion from the real issue.
Mike Dunford
> unre...@hotmail.com (MEC) wrote in
> news:c240c53.02082...@posting.google.com:
[snip]
>> a) What point was Dawkins trying to make when
>> he described WEASEL?
>
> I cannot read Dawkins's mind.
That is understandable, and it is entirely possible that even if you
could read his mind, you might not want to. Fortunately, however,
there is an alternative method available to the psi-impaired
among us who might still want to figure out what point he had in mind
here. It's a revolutionary concept, but perhaps you might just want
to try reading his book, instead. I would suggest that you look at
pages 43-49 (or pages 46-49, if the first selection exceeds your
attention span) of _The Blind Watchmaker_.
Here's the key passage:
What matters is the difference between the time taken
by _cumulative_ selection and the time which the same
computer, working at the same rate, would take to
reach the target phrase if it were forced to use the
other procedure of _single step selection_....
There is a big difference, then, between cumulative
selection (in which each improvement, however slight, is
used as a basis for future building), and single-step
selection (in which each new 'try' is a fresh one). If
evolutionary process had to rely on single-step selection
it could never have got anywhere.
_Blind Watchmaker_ p.49
As just about everyone has unsuccessfully tried to point out to you,
this example does not claim to be an illustration, a model, or a
simulation of natural selection. It is, as Dawkins points out, simply
an example of the power of cumulative selection.
> But what he DID DO was use his
> WEASEL simulation to demonstrate natural selection's: (1)
> "success" and (2) speed.
No, Walter, he did not. He (correctly) points out that natural
selection acts by using cumulative selection, but he did not claim
that the WEASEL program was an example of natural selection. It was
simply a very good, clear illustration of the difference between the
odds of something happening through single-step selection and the
odds of the same thing happening using cumulative selection instead.
Again, nobody but you thinks that the WEASEL program illustrates,
demonstrates, or simulates natural selection in any way shape or
form.
> BOTH of those are illusions dismantled in my book.
Both of those strawmen might be dismantled, but they are just that --
strawmen. You certainly did not dismantle Dawkins' argument, because
he didn't make that argument.
[snip]
--Mike Dunford
--
On two occasions I have been asked [by members of Parliament],
'Pray, Mr. Babbage, if you put into the machine wrong figures, will
the right answers come out?' I am not able rightly to apprehend the
kind of confusion of ideas that could provoke such a question.
--Charles Babbage
Adam Marczyk
> "Adam Marczyk" <ebon...@hotmailNOTexcite.com> wrote in
> news:umbasm1...@corp.supernews.com:
>
> > .... The WEASEL program was not *ever* intended to be a
> > simulation of evolution. Never.
>
> Dawkins's WEASEL program was used to demonstrate natural selection's: (1)
> "success", and (2) speed.
No. You continue to be wrong. The WEASEL program demonstrates nothing about
*natural* selection, per se, and it was never claimed to by anyone (except
misguided creationists such as yourself). What it does demonstrate is the
success and speed of *cumulative* selection - a more general process which,
unlike NS, may have a target chosen ahead of time - as compared to single-step
guessing.
Adam Marczyk
> unre...@hotmail.com (MEC) wrote in
> news:c240c53.02082...@posting.google.com:
>
> > .... As Ian [Musgrave] and others have been trying
> > to tell you ...
>
> Musgrave said "SIZE MATTERS". He is wrong to leave the argument there. In
> Dawkins's simulation, "size matters" IN FAVOR OF EVOLUTION! Musgrave has
> no grounds for assaulting me on this. He is repeatedly misrepresenting my
> argument (by gross omission).
>
> Musgrave's claims about "Haldane's Dilemma" are false, because they arise
> from his false definition of Haldane's Dilemma. Again he is
> misrepresenting my argument (by gross omission).
You must be incredibly arrogant to claim he's misrepresenting you by gross
omission when you delete his entire posts and argue against what you claim he
must be saying, rather than allowing his words to speak for themselves.
> > a) What point was Dawkins trying to make when
> > he described WEASEL?
>
> I cannot read Dawkins's mind. But what he DID DO was use his WEASEL
> simulation to demonstrate natural selection's: (1) "success" and (2) speed.
I concur with Mike Dunford. While you cannot read Dawkins' mind, you *can* read
Dawkins' book, in which he clearly and explicitly states the purpose of the
WEASEL program. That purpose is to show the power of cumulative selection over
random single-step guessing. You'll note I said cumulative selection, and not
natural selection, because the WEASEL program is intended to demonstrate the
former and *not* the latter (your repeated erroneous claims to the contrary).
Bobby D. Bryant
with this .sig:
> On two occasions I have been asked [by members of Parliament],
> 'Pray, Mr. Babbage, if you put into the machine wrong figures,
> will the right answers come out?' I am not able rightly to
> apprehend the kind of confusion of ideas that could provoke such a
> question.
> --Charles Babbage
How apropos, when attached to a post explaining ReMine's confusion
of ideas!
Bobby Bryant
Austin, Texas
Mike Dunford
news:ak4ol2$sf$1...@geraldo.cc.utexas.edu:
I use a random .sig generator -- it picks a new entry from my
quotes.txt file every couple of minutes. There are times such as this
when I think that it's posessed by some sort of appropriateness
demon.
--Mike Dunford
--
If it looks like a duck, and quacks like a duck, we have at least to
consider the possibility that we have a small aquatic bird of the
family anatidae on our hands.
--Douglas Adams
Roy Thearle
> > .... By continually attributing this program and its
> > default parameters to Dawkins, Mr ReMine ...
> > misprepresents Dawkins ...
>
> Wilson has shown no such misrepresentation.
Are you
(i) denying that you attributed the program as being by Dawkins?
(ii) saying that you don't think it was a misrepresentation?
(iii) equivocating by pointing out that it was someone other than
Wilson who raised the issue?
Anyway, he doesn't need to show anything since it's obvious to anyone
reading these discussions.
You wrote:
"Trying to compare the substitution rate
in a population of [N=100] individuals
with the substitution rate in a population
of between 10,000 to 100,000 individuals
is a pretty big blunder to make"
(from Musgrave's post, brackets mine,
N=100 from Dawkins's simulation)
You also wrote:
To begin my research, I had first sent Dawkins a check,
requesting a copy of his program. He kindly
wrote back, and returned the
check, saying the simulation had been "lost."
So the N=100* wasn't actually from Dawkins's simulation as you
claimed.
Roy
*later changed without apology or explanation to N>=100
Noelie S. Alito
> "Adam Marczyk" <ebon...@hotmailNOTexcite.com> wrote in
> news:umbasm1...@corp.supernews.com:
>
> > .... The WEASEL program was not *ever* intended to be a
> > simulation of evolution. Never.
>
> Dawkins's WEASEL program was used to demonstrate natural selection's: (1)
> "success", and (2) speed.
All right, Quotemaster: Give us an EXACT QUOTE where Dawkins
makes this claim. Quit misrepresenting him.
> Both are illusions my book dismantles.
Man, get some medication....
Noelie
--
When you argue with a compulsive nutjob, chances are he is doing the same.
Richard Uhrich
> "Walter ReMine" <sci...@nospam.net> wrote in message
> news:Xns9272EB134B8D3...@206.191.138.115...
>
>>"Adam Marczyk" <ebon...@hotmailNOTexcite.com> wrote in
>>news:umbasm1...@corp.supernews.com:
>>
>>
>>>.... The WEASEL program was not *ever* intended to be a
>>>simulation of evolution. Never.
>>>
>>Dawkins's WEASEL program was used to demonstrate natural selection's: (1)
>>"success", and (2) speed.
>>
>
>
> All right, Quotemaster: Give us an EXACT QUOTE where Dawkins
> makes this claim. Quit misrepresenting him.
>
Put up or shut up, huh? Good idea! Let's end the quibbling and
distortions. (I too didn't see claims in "Watchmaker" about this program
other than that it demonstrates the astronomical odds against a chance
one-time event are irrelevant when mutation and selection are introduced.)
<snip>
John Stockwell
>> > he described WEASEL?
>>
>> I cannot read Dawkins's mind. But what he DID DO was use his WEASEL
>> simulation to demonstrate natural selection's: (1) "success" and (2) speed.
>
>I concur with Mike Dunford. While you cannot read Dawkins' mind, you *can* read
>Dawkins' book, in which he clearly and explicitly states the purpose of the
>WEASEL program. That purpose is to show the power of cumulative selection over
>random single-step guessing. You'll note I said cumulative selection, and not
>natural selection, because the WEASEL program is intended to demonstrate the
>former and *not* the latter (your repeated erroneous claims to the contrary).
Just to jump in,
The success of genetic algorithms for solving specific problems only occurs
if a specific fitness function is known and applied. The point is, however,
that Darwinian evolution is *not* a search algorithm. There is no predetermined
fitness function that a solution is being optimized under.
Indeed, as Dembski rightly points out, the "no free lunch theorems" show
that genetic algorithms are no better than random search, in general,
as an optimising tool---if we are considering genetic algorithms as
searching algorithms.
Interestingly, I believe that history will hail Dembski as being the person
who finally has dealt "intelligent design" its death blow. Why? So far
nobody has shown that additional mechanisms to those of mutation, genetic
drift and natural selection are necessary for the origin of species.
Furthermore, nobody has shown that there are any alternative mechanisms
by which species could arise, such as "design". This means that some
sort of Darwinian mechanism is the *only* available mechanism for the origin of
species. So, even if we want to believe that the "designer" is using
evolution as "his/her/or its" mechanism for creating species, the NFL
theorems prevent this from being a mechanism for "design" that is any
more efficient than some sort of random search,
The inescapable conclusions are that there is nothing like "complex
specified information" (either of Dembski's or any other criterion) in
biology, except in the eye of the beholder. Design is dead, killed by
the Discovery Institute.
-John
PS. Regarding the "biotic message" perhaps some avant guard geologist
will write a book called the "petrographic message" where in he or
she translates the hidden messages written in graphic granite.
--
John Stockwell | jo...@dix.Mines.EDU
Center for Wave Phenomena (The Home of Seismic Un*x)
Colorado School of Mines
Golden, CO 80401 | http://www.cwp.mines.edu/cwpcodes
voice: (303) 273-3049
Andy Groves
> unre...@hotmail.com (MEC) wrote in
> news:c240c53.02082...@posting.google.com:
>
> > .... As Ian [Musgrave] and others have been trying
> > to tell you ...
>
> Musgrave said "SIZE MATTERS". He is wrong to leave the argument there. In
> Dawkins's simulation, "size matters" IN FAVOR OF EVOLUTION! Musgrave has
> no grounds for assaulting me on this. He is repeatedly misrepresenting my
> argument (by gross omission).
>
> Musgrave's claims about "Haldane's Dilemma" are false, because they arise
> from his false definition of Haldane's Dilemma. Again he is
> misrepresenting my argument (by gross omission).
>
> > a) What point was Dawkins trying to make when
> > he described WEASEL?
>
> I cannot read Dawkins's mind. But what he DID DO was use his WEASEL
> simulation to demonstrate natural selection's: (1) "success" and (2) speed.
Dawkins's weasel simulation did *not* simulate natural selection. He
made that point **explicitly clear** in his book. Why can you not
grasp this simple fact?
Andy
Bobby D. Bryant
> "Walter ReMine" <sci...@nospam.net> wrote in message
> news:Xns9272EB134B8D3...@206.191.138.115...
>> "Adam Marczyk" <ebon...@hotmailNOTexcite.com> wrote in
>> news:umbasm1...@corp.supernews.com:
>>
>> > .... The WEASEL program was not *ever* intended to be a
>> > simulation of evolution. Never.
>>
>> Dawkins's WEASEL program was used to demonstrate natural
>> selection's: (1) "success", and (2) speed.
>
>
> All right, Quotemaster: Give us an EXACT QUOTE where Dawkins
> makes this claim. Quit misrepresenting him.
I'm just glad it wasn't:
Dawkins's WEASEL program was used to demonstrate natural
selection's: (1) [success], and (2) [speed]
Bobby Bryant
Austin, Texas
petehope
> On Sat, 10 Aug 2002 17:29:20 -0600, Ian Musgrave & Peta O'Donohue
> wrote:
>
> > Dawkins program was not designed to be a simulation of natural
> > selection. Dawkins says this explicitly and mentions some of the
> > reasons that it isn't, and what a real simulation would entail.
> >
> > What it _is_ is a demonstration of the efficacy of selection over
> > random search (which it does quite admirably).
>
> I've posted this before, but new lurkers may want to check out
> http://www.cs.utexas.edu/users/bdbryant/talk-origins/ga-on-tsp/index.html
>
> The page has some screenshots of a GA that I hacked to show
> side-by-side shots of the progress of selected search vs. random
> search on the Travelling Salesman Problem.
>
> It's not a very good model of biology either, but it has a nice
> graphical output, and unlike WEASEL it does not rely on a
> pre-specified answer. (The "fitness" rule is simlply shorter path =
> better.)
>
>
> Also, though it may not be possible to map Haldane's Delimma onto
> what this solver does, I _have_ run it with lots of different
> population sizes, and the general rule is that larger populations
> find better solutions.
>
> Bobby Bryant
> Austin, Texas
I love it! I think it also shows that the more complex the system is
(here: more cities), the better the GA tends to do compared to the
random search. Now, considering how complex biological problems are,
random search wouldn't have a prayer. I don't know how ID people can
keep claiming that GA (natural selection) couldn't come up with better
solutions! It's so obvious, you need to be stupid, extremely
narrow-minded, or just plain evil and politically motivated to not
(want to) see it.
David Sienkiewicz
> > But what he DID DO was use his
> > WEASEL simulation to demonstrate natural selection's: (1)
> > "success" and (2) speed.
>
> No, Walter, he did not. He (correctly) points out that natural
> selection acts by using cumulative selection, but he did not claim
> that the WEASEL program was an example of natural selection. It was
> simply a very good, clear illustration of the difference between the
> odds of something happening through single-step selection and the
> odds of the same thing happening using cumulative selection instead.
>
> Again, nobody but you thinks that the WEASEL program illustrates,
> demonstrates, or simulates natural selection in any way shape or
> form.
>
> > BOTH of those are illusions dismantled in my book.
>
> Both of those strawmen might be dismantled, but they are just that --
> strawmen. You certainly did not dismantle Dawkins' argument, because
> he didn't make that argument.
> [snip]
>
> --Mike Dunford
I'm thinking that Walter shows up every now and then because he wants
to generate some controversy and sell his book.
But it seems to me that every time he does that, we get more reasons
NOT to buy the book.
Walter ReMine
> .... What [Dawkins's WEASEL program] does demonstrate
> is the success and speed of *cumulative* selection
I dismantle it's "success" and speed as illusions.
Brian M. Scott
More dishonest snipping. What Adam actually said is that It
demonstrates the success and speed of cumulative selection
_as_compared_to_single-step_guessing_. And no, you do not demonstrate
that its relative speed and success _in_this_comparison_ are
illusions. First, they aren't illusions, as anyone with a WEASEL-like
program can easily prove for himself. Secondly, you don't even
address this comparison, since you do not acknowledge that it is the
comparison actually being made. On the available evidence you appear
not even to understand it, though it's hard to imagine that anyone
could be so dense.