Math and Universe
Seth Russell
> Mathematics is useful to scientists, precisely because mathematics
> *is* epistemology, the theory of knowledge.
>
> That other subject, that entirely bogus branch of philosophy usually
> known as 'epistemology' is a pseudo-theory of pseudo-knowledge. As
> one might expect, it has not proven significantly useful to
> scientists.
How do you know that?
Seth Russell
http://RobustAi.Net/Ai/SymKnow.Htm
Neil W Rickert
>Neil W Rickert wrote:
>> Mathematics is useful to scientists, precisely because mathematics
>> *is* epistemology, the theory of knowledge.
>> That other subject, that entirely bogus branch of philosophy usually
>> known as 'epistemology' is a pseudo-theory of pseudo-knowledge. As
>> one might expect, it has not proven significantly useful to
>> scientists.
>How do you know that?
Which part?
The mentioned branch of philosophy concerns itself with "justified
true belief". I claim it is a pseudo-theory because it fails to give
a convincing account of justification, truth or belief. I claim it
is about pseudo-knowledge, because I deny that knowledge is justified
true belief. Note, for example, that laboratory classes are
considered an essential part of a scientist's education, for a
scientists knowledge has to do with abilities rather than beliefs.
Check Feynam's chapter "O Americano, Outra Vez!" in his book "Surely
You're Joking, Mr Feynman" for an example of what happens when
somebody acquires the beliefs but not the abilities.
I claim that mathematics is the real epistemology, because it acts as
a meta-theory of science (and even of common-sense knowledge),
providing the underlying basis for methods ranging from probability
and statistics to arithmetic to calculus. More generally, our
knowledge is in our ability to find systematic ways of dealing with
the problems that reality imposes upon us. Mathematics is the theory
of systematicity (or the theory of pattern), and this is what makes
it the real theory of knowledge.
james d. hunter
>
> Seth Russell <se...@halcyon.com> writes:
> >Neil W Rickert wrote:
>
> >> Mathematics is useful to scientists, precisely because mathematics
> >> *is* epistemology, the theory of knowledge.
>
> >> That other subject, that entirely bogus branch of philosophy usually
> >> known as 'epistemology' is a pseudo-theory of pseudo-knowledge. As
> >> one might expect, it has not proven significantly useful to
> >> scientists.
>
> >How do you know that?
>
> Which part?
>
> The mentioned branch of philosophy concerns itself with "justified
> true belief". I claim it is a pseudo-theory because it fails to give
> a convincing account of justification, truth or belief. I claim it
> is about pseudo-knowledge, because I deny that knowledge is justified
> true belief. Note, for example, that laboratory classes are
> considered an essential part of a scientist's education, for a
> scientists knowledge has to do with abilities rather than beliefs.
> Check Feynam's chapter "O Americano, Outra Vez!" in his book "Surely
> You're Joking, Mr Feynman" for an example of what happens when
> somebody acquires the beliefs but not the abilities.
has an unjustified theory of truth, but they use pieces of it anyway
for what's it's worth.
Chris Hooley
[In a previous post]
> Mathematics is useful to scientists, precisely because mathematics
> *is* epistemology, the theory of knowledge.
No one questions this limited definition? Epistemology is not only
theory, it is also the study of the grounds of knowledge. It is (or can
be) one of those endeavors that profits us most through its process, rather
than its product. It is the question, "And on what grounds do you know
*that*....", like the child's repeated question, "But, why...." It is the
questioning that matters.
To say that "mathematics *is* epistemology," seems like saying that
"Christianity *is* theology." It sounds to me like a Mid-eighteenth
Century Englishman, thumbs hooked in his trousers, saying, "The British
Empire *is* the world, My Good Man."
> That other subject, that entirely bogus branch of philosophy usually
> known as 'epistemology' is a pseudo-theory of pseudo-knowledge. As
> one might expect, it has not proven significantly useful to
> scientists.
>The mentioned branch of philosophy concerns itself with "justified
>true belief". I claim it is a pseudo-theory because it fails to give
>a convincing account of justification, truth or belief. I claim it
>is about pseudo-knowledge, because I deny that knowledge is justified
>true belief.
Again, Neil, I am surprised at this usage. I was not aware that
epistemology was a single theory. Perhaps this is a change that has
occured since I was in college (a long time ago, I grant).
Epistemology serves my thinking. Like a court jester, it reminds me to
question the validity and limits of knowledge. It cavorts about,
unimpressed with my rational productions. "Aha, think your so smart,
Chris! And on what grounds do you think you know *that*..."
I feel that some of the most important breakthoughs in AI research lie
beneath "what we think we know," so I appreciated this sense of
epistemology.
Regards,
Chris
Jerry Hull
wrote:
>Seth Russell <se...@halcyon.com> writes:
>>Neil W Rickert wrote:
>
>>> Mathematics is useful to scientists, precisely because mathematics
>>> *is* epistemology, the theory of knowledge.
>
>>> That other subject, that entirely bogus branch of philosophy usually
>>> known as 'epistemology' is a pseudo-theory of pseudo-knowledge. As
>>> one might expect, it has not proven significantly useful to
>>> scientists.
>
>>How do you know that?
>
>Which part?
>
>The mentioned branch of philosophy concerns itself with "justified
>true belief". I claim it is a pseudo-theory because it fails to give
>a convincing account of justification, truth or belief. I claim it
>is about pseudo-knowledge, because I deny that knowledge is justified
>true belief.
>I claim that mathematics is the real epistemology, because it acts as
>a meta-theory of science (and even of common-sense knowledge),
>providing the underlying basis for methods ranging from probability
>and statistics to arithmetic to calculus. More generally, our
>knowledge is in our ability to find systematic ways of dealing with
>the problems that reality imposes upon us. Mathematics is the theory
>of systematicity (or the theory of pattern), and this is what makes
>it the real theory of knowledge.
The obvious problem with this is that, if mathematics is the "real"
epistemology, it follows that all we "really" know are mathematical
entities and mathematical properties. But science supposes we are
able to know other things, the things NUMBERED by mathematics: e.g.
statial location, energy, mass, speed, density, chemical composition,
taste, color, odor, feel, smell, &c.
So is everything "really" a mathematical entity or property? Prima
facie, empirical properties such as those listed above are NOT such,
however useful math may be in explicating their features and
relationships. So either you have the mother of all reductivisms, or
some serious backpedaling to do.
--
Jer
"When you are at sea, keep clear of the land",
Publilius Syrus
Neil W Rickert
>On 4 Dec 1999 14:44:02 -0600, Neil W Rickert <ricke...@cs.niu.edu> wrote:
>[In a previous post]
>> Mathematics is useful to scientists, precisely because mathematics
>> *is* epistemology, the theory of knowledge.
>No one questions this limited definition?
They probably think my view is too nutty to be worth responding.
> Epistemology is not only
>theory, it is also the study of the grounds of knowledge.
Mathematics certainly has been important for scientists in
establishing grounds for knowledge. Nothing comparable has come from
that branch of philosphy known as epistemology.
> It is (or can
>be) one of those endeavors that profits us most through its process, rather
>than its product.
And indeed, it is the process of mathematics that I was referring
to.
> It is the question, "And on what grounds do you know
>*that*....", like the child's repeated question, "But, why...." It is the
>questioning that matters.
As one example, the mathematical theory of probability is used to
justify statistical methods used in some of the sciences.
You won't find much in the writing of philosophers that helps here.
In "Logical Foundations of Probability", Carnap admits that he
methods cannot account for laws such as the theory of relativity, and
that his theory of confirmation provides a degreee of confirmation of
exactly 0% for such laws.
>To say that "mathematics *is* epistemology," seems like saying that
>"Christianity *is* theology." It sounds to me like a Mid-eighteenth
>Century Englishman, thumbs hooked in his trousers, saying, "The British
>Empire *is* the world, My Good Man."
I did give some reasons for my assertion in my response to Seth
Russell.
>> That other subject, that entirely bogus branch of philosophy usually
>> known as 'epistemology' is a pseudo-theory of pseudo-knowledge. As
>> one might expect, it has not proven significantly useful to
>> scientists.
>>The mentioned branch of philosophy concerns itself with "justified
>>true belief". I claim it is a pseudo-theory because it fails to give
>>a convincing account of justification, truth or belief. I claim it
>>is about pseudo-knowledge, because I deny that knowledge is justified
>>true belief.
>Again, Neil, I am surprised at this usage. I was not aware that
>epistemology was a single theory. Perhaps this is a change that has
>occured since I was in college (a long time ago, I grant).
You are right. It is not even a single theory.
>Epistemology serves my thinking. Like a court jester, it reminds me to
>question the validity and limits of knowledge. It cavorts about,
>unimpressed with my rational productions. "Aha, think your so smart,
>Chris! And on what grounds do you think you know *that*..."
Why is it, do you think, that physical scientists have found no use
at all for for epistemology, except perhaps to mention Popper (whom
they mostly have not read) when questioning something like
astrology?
>I feel that some of the most important breakthoughs in AI research lie
>beneath "what we think we know," so I appreciated this sense of
>epistemology.
It is almost 50 years since Turing wrote his "Mind" paper, and AI so
far has been quite unsuccessful at using epistemology (so called),
although it has successfully used mathematics.
Neil W Rickert
>On 4 Dec 1999 14:44:02 -0600, Neil W Rickert <ricke...@cs.niu.edu>
>wrote:
>>I claim that mathematics is the real epistemology, because it acts as
>>a meta-theory of science (and even of common-sense knowledge),
>>providing the underlying basis for methods ranging from probability
>>and statistics to arithmetic to calculus. More generally, our
>>knowledge is in our ability to find systematic ways of dealing with
>>the problems that reality imposes upon us. Mathematics is the theory
>>of systematicity (or the theory of pattern), and this is what makes
>>it the real theory of knowledge.
>The obvious problem with this is that, if mathematics is the "real"
>epistemology, it follows that all we "really" know are mathematical
>entities and mathematical properties. But science supposes we are
>able to know other things, the things NUMBERED by mathematics: e.g.
>statial location, energy, mass, speed, density, chemical composition,
>taste, color, odor, feel, smell, &c.
You only underscore my point about the uselessness of that branch of
philosophy known as epistemology. Mathematics has been one of the
most useful fields of study, as even epistemologists (so called) will
admit. Yet epistemologists (so called) have failed to account for
mathematics. Mathematics seems, at least on some accounts, to
consist of beliefs which are analytic truths and therefore which
epistemologists (so called) take as having no content. And the
entities that mathematics talks about appear to exist only in a
mythical world of platonic forms. So if epistemology (so called) is
even close to being a useful discipline, then mathematics ought to be
the most useless of all fields of study.
It seems to me that you are making an excellent case for the bogosity
of epistemology (so called).
>So is everything "really" a mathematical entity or property? Prima
>facie, empirical properties such as those listed above are NOT such,
>however useful math may be in explicating their features and
>relationships. So either you have the mother of all reductivisms, or
>some serious backpedaling to do.
This is the sort of confusion that one could expect to come from people
indoctrinated into the religion which is philosophy.
Mathematics isn't really the study of mathematical entities, so it
doesn't matter whether or not those entities actually exist. And the
content of mathematics is not the theorems, so it doesn't matter
whether those theorems are about anything or are only about imaginary
entities. Mathematics is really about methodologies, and the content
of mathematics is the useful methodologies, many of which have been
adopted by science. Mathematical entities are only invented so that
the methodologies can be thoroughly tested in a manner which is
independent of the details of the real situations where they will be
used.
Seth Russell
> >> Mathematics is useful to scientists, precisely because mathematics
> >> *is* epistemology, the theory of knowledge.
> >> That other subject, that entirely bogus branch of philosophy usually
> >> known as 'epistemology' is a pseudo-theory of pseudo-knowledge. As
> >> one might expect, it has not proven significantly useful to
> >> scientists.
Well, if you want to adopt only the highest standard of verification,
then mathematics and the abilities of science are what we know and
how we know it. And i think that is pretty much the definition of
epistemology. So basically I think i see what you are saying. However,
at its best, this web of knowledge is a small representation (reflection in
our subjective minds) of all that actually is. How can this same web of
math/science/abilities understand itself without recourse to something
that is ~not~ math/science/abilities? Otherwise don't we get into the
same reflexive paradox that lead mathematician and logicians to say
things like "you cannot model a language without stepping outside of
the language" ?
Seth Russell
Http://RobustAi.Net/Ai/SymKnow.htm
Neil W Rickert
>Neil W Rickert wrote:
>> >> Mathematics is useful to scientists, precisely because mathematics
>> >> *is* epistemology, the theory of knowledge.
>> >> That other subject, that entirely bogus branch of philosophy usually
>> >> known as 'epistemology' is a pseudo-theory of pseudo-knowledge. As
>> >> one might expect, it has not proven significantly useful to
>> >> scientists.
>Well, if you want to adopt only the highest standard of verification,
>then mathematics and the abilities of science are what we know and
>how we know it.
I'm not sure why you are picking out this one paragraph, and not
commenting on the things I have said more recently, such as in my
reply to your earlier posting.
It is one of my criticisms of epistemolgy (the branch of philosophy,
so called), that it vastly overrates the importance of verification.
If you think that I am adopting excessively high standards of
verification, then you have entirely missed the point.
Incidently, why do you send me an email copy of your postings? And,
more importantly, why do you edit the address to which you send
them? I wish you would not do that. I am deliberately using a
posting address that I can filter into a separate mailbox from my
regular email.
Seth Russell
> I'm not sure why you are picking out this one paragraph, and not
> commenting on the things I have said more recently, such as in my
> reply to your earlier posting.
Your earlier posting elaborated and informed, but I had no specific
reaction to it.
> It is one of my criticisms of epistemolgy (the branch of philosophy,
> so called), that it vastly overrates the importance of verification.
> If you think that I am adopting excessively high standards of
> verification, then you have entirely missed the point.
I though that math provided exact and unambiguous match points
for the purpose of verification. How can you have a science of
reliably knowing, without verification? Could you elaborate?
> Incidently, why do you send me an email copy of your postings? And,
> more importantly, why do you edit the address to which you send
> them? I wish you would not do that. I am deliberately using a
> posting address that I can filter into a separate mailbox from my
> regular email.
Sorry. I use the 'reply all' mechanism in Netscape as a strategy
to encourage private dialogue. It frequently works for me,
just as often it annoys others. Now that I know it annoys you,
I'll respond only to your posts in the newsgroups.
Seth Russell
Neil W Rickert
>Neil W Rickert wrote:
>> I'm not sure why you are picking out this one paragraph, and not
>> commenting on the things I have said more recently, such as in my
>> reply to your earlier posting.
>Your earlier posting elaborated and informed, but I had no specific
>reaction to it.
>> It is one of my criticisms of epistemolgy (the branch of philosophy,
>> so called), that it vastly overrates the importance of verification.
>> If you think that I am adopting excessively high standards of
>> verification, then you have entirely missed the point.
>I though that math provided exact and unambiguous match points
>for the purpose of verification.
Sure.
> How can you have a science of
>reliably knowing, without verification? Could you elaborate?
You need a way of judging success or usefulness. But it is not
simply a matter of verification.
When confronted with a question about mathematical truths, one of my
professors (S. Kakutani) used to ask:
Pick a number. Square it. Is that a theorem?
His point was this: If you start with a large integer n, you can
square it, to yield the integer m = n^2. Then the assertion
m is a perfect square
where the actual value of m is subtituted, is a true statement. If
the original number n was large enough, then very likely this is the
first time in the history of the human race that the truth of this
statement has ever been asserted. Yet no mathematical journal would
publish this as a research result, and no mathematician would
consider it an important breakthrough.
The point I am trying to make is that there is very little knowledge
in a true statement. "Justified true belief" is not what a
mathematician means by "knowledge". Nor is it what a physicist
means, or probably what any physical scientist means by "knowledge".
Let me give another example.
Let's suppose that a is a 550 digit prime number. Let's also suppose
that b is a 450 digit prime number. Let's suppose that when you
multiply them, you get c, which is then a 999 or 1000 digit composite
number.
If you submitted an article entitled "a * b = c", but with no proof,
your article would be rejected by any respectable math journal. If
you submitted it with a proof, and the proof consisted of multiplying
a and b, and showing that the answer is c, then your article would be
rejected by any respectable math journal. If, however, you submitted
it with a proof that started with c, and was cleverly able to come up
with a factorization without any prior knowledge of a and b, without
using anything special that makes factoring c unusually easy compared
to other numbers of a similar magnitude, and without any reliance on
lucky guesses, then the article would likely be published, and there
would probably be many mathematicians who studied that article. Note
that current the state of the art of factorization is not anywhere
near being able to deal with 1000 digit numbers that are products of
two primes.
What is being looked for is not the mathematical truth, but the
method used to unearth that truth. The fact that the method produced
a true statement is important in judging the usefulness of the
method. But the truth itself is not the mathematical knowledge.
Let me give another example. Suppose that you had a black box.
Every time you turned the handle, out came a previously unknown
truth. We would not count the memorization of that output as
significant knowledge. However, we would be very interested in the
blueprint for building such a black box, and the ability to construct
such a black box would be considered knowledge. We value science not
because it has given us previously unknown truths, but because it has
given us the blueprints for many black boxes.
The idea that knowledge = justified true belief might be just what is
required in a theological seminary. But it is wrong for mathematics,
it is wrong for science, and it is surely wrong for ordinary
knowledge. Perhaps I should add that it is wrong for AI, and this is
why projects such as CYC, based on that bogus concept of knowledge,
will never succeed.
james d. hunter
In science it is exactly a -justified- by experiment belief.
The "knowledge" in physical science is identical to the "knowledge"
in math. They are both mathematical theories. Like the old saying
goes: When you go to a ballgame, you need a program to tell the
players apart.
Jerry Hull
wrote:
>Seth Russell <se...@halcyon.com> writes:
>> How can you have a science of
>>reliably knowing, without verification? Could you elaborate?
>
>You need a way of judging success or usefulness. But it is not
>simply a matter of verification.
>
>When confronted with a question about mathematical truths, one of my
>professors (S. Kakutani) used to ask:
>
> Pick a number. Square it. Is that a theorem?
>
>His point was this: If you start with a large integer n, you can
>square it, to yield the integer m = n^2. Then the assertion
>
> m is a perfect square
>
>where the actual value of m is subtituted, is a true statement. If
>the original number n was large enough, then very likely this is the
>first time in the history of the human race that the truth of this
>statement has ever been asserted. Yet no mathematical journal would
>publish this as a research result, and no mathematician would
>consider it an important breakthrough.
>
>The point I am trying to make is that there is very little knowledge
>in a true statement. "Justified true belief" is not what a
>mathematician means by "knowledge". Nor is it what a physicist
>means, or probably what any physical scientist means by "knowledge".
Your example hardly establishes this point. All you've shown is that
not all knowledge is SIGNIFICANT knowledge. This does not imply that
some account of knowledge in terms of justified belief cannot be true.
>Let me give another example.
>
>Let's suppose that a is a 550 digit prime number. Let's also suppose
>that b is a 450 digit prime number. Let's suppose that when you
>multiply them, you get c, which is then a 999 or 1000 digit composite
>number.
>
>If you submitted an article entitled "a * b = c", but with no proof,
>your article would be rejected by any respectable math journal. If
>you submitted it with a proof, and the proof consisted of multiplying
>a and b, and showing that the answer is c, then your article would be
>rejected by any respectable math journal. If, however, you submitted
>it with a proof that started with c, and was cleverly able to come up
>with a factorization without any prior knowledge of a and b, without
>using anything special that makes factoring c unusually easy compared
>to other numbers of a similar magnitude, and without any reliance on
>lucky guesses, then the article would likely be published, and there
>would probably be many mathematicians who studied that article.
And all this shows is that we grade knowledge with respect to its
significance, and showing a way to do something that has not been
previously discovered is rather more significant than things that can
be easily accomplished by anybody.
>What is being looked for is not the mathematical truth, but the
>method used to unearth that truth. The fact that the method produced
>a true statement is important in judging the usefulness of the
>method. But the truth itself is not the mathematical knowledge.
This is just confused. That such-and-such a procedure guarantees
results in a finite number of steps is itself a TRUTH of mathematics.
Again, you fail to separate truth from SIGNIFICANT truth. But who
every said that all knowledge has to be significant?
>Let me give another example. Suppose that you had a black box.
>Every time you turned the handle, out came a previously unknown
>truth. We would not count the memorization of that output as
>significant knowledge.
This introduces several more ambiguities. How do we know that what
the box is emitting is true? Do we have an independent effective
procedure for testing its claims? Because if the box is emitting
putative theorems (without accompanying demonstrations), there is no
way to test if it is accurate. On the other hand, if it turns out
that we can (e.g. empirically) verify what the box is emitting, and on
the basis of its past success have gained confidence in its results,
then it would indeed make sense to mark its output down, as we would
for any fortune teller with a good track record.
> However, we would be very interested in the
>blueprint for building such a black box, and the ability to construct
>such a black box would be considered knowledge. We value science not
>because it has given us previously unknown truths, but because it has
>given us the blueprints for many black boxes.
But the boxes are ONLY valuable as the MEANS to the truths they
generate. Who cares about the boxes who doesn't care about truth?
>The idea that knowledge = justified true belief might be just what is
>required in a theological seminary. But it is wrong for mathematics,
>it is wrong for science, and it is surely wrong for ordinary
>knowledge. Perhaps I should add that it is wrong for AI, and this is
>why projects such as CYC, based on that bogus concept of knowledge,
>will never succeed.
It is hard to imagine ANY interesting intelligence, machine or
human-like, that did not (a) have representations (models) of itself
and the world, and (b) have some means of testing the adequacy of
those representations. (a) give you belief, and (b) gives you
justification. Since no belief is justified if it is not true (this
is MY wrinkle on the formula), that gives you justified true belief.
I am ignorant about CYC, so you may be entirely correct where it is
concerned.
Neil W Rickert
>Neil W Rickert <ricke...@cs.niu.edu> wrote:
>>Seth Russell <se...@halcyon.com> writes:
>>The point I am trying to make is that there is very little knowledge
>>in a true statement. "Justified true belief" is not what a
>>mathematician means by "knowledge". Nor is it what a physicist
>>means, or probably what any physical scientist means by "knowledge".
>Your example hardly establishes this point. All you've shown is that
>not all knowledge is SIGNIFICANT knowledge.
It is surely one of the faults of epistemology (so called) that it
makes no such distinction.
> This does not imply that
>some account of knowledge in terms of justified belief cannot be true.
I am not claiming that the claims of epistemology (so called) are
untrue. They are surely analytically true but worthless dogmas,
believed religiously by the adherents of that religion. They are
analytically true because epistemologists distort the meaning of
"knowledge" so as to make them true.
>>Let me give another example. Suppose that you had a black box.
>>Every time you turned the handle, out came a previously unknown
>>truth. We would not count the memorization of that output as
>>significant knowledge.
>This introduces several more ambiguities. How do we know that what
>the box is emitting is true? Do we have an independent effective
>procedure for testing its claims? Because if the box is emitting
>putative theorems (without accompanying demonstrations), there is no
>way to test if it is accurate. On the other hand, if it turns out
>that we can (e.g. empirically) verify what the box is emitting, and on
>the basis of its past success have gained confidence in its results,
>then it would indeed make sense to mark its output down, as we would
>for any fortune teller with a good track record.
>> However, we would be very interested in the
>>blueprint for building such a black box, and the ability to construct
>>such a black box would be considered knowledge. We value science not
>>because it has given us previously unknown truths, but because it has
>>given us the blueprints for many black boxes.
>But the boxes are ONLY valuable as the MEANS to the truths they
>generate. Who cares about the boxes who doesn't care about truth?
Why do philosophers make such a religion out of their silly ideas
about 'truth'.
Hooke's law is analytic, if taken to be an assertion about ideal
perfectly elastic bodies. If taken to be about actual physical
things, then Hooke's law is false, and it is well known to be false.
Boyle's law is analytic if taken to be an assertion about ideal
gasses. If taken to be about actual gasses, it is false, and it is
well known to be false.
Ohm's law is analyic if resistance is taken to be an abstract
property of a circuit element defined so as to satisfy Ohm's law. If
resistance is taken to be an intrinsic property of a circuit element,
the Ohm's law is false, and is known to be false.
Much the same can be said about all of our scientific laws.
Depending on how we interpret them, they are either analytic truths
and presumably therefore contain no knowledge, or they are false (and
known to be false) and therefore contain no knowledge.
According to Hull's ideas about epistemology, there is presumably no such
thing as scientific knowledge.
Or, as I would prefer to say it, that branch of philosophy known as
epistemology is dogmatic religion.
>>The idea that knowledge = justified true belief might be just what is
>>required in a theological seminary. But it is wrong for mathematics,
>>it is wrong for science, and it is surely wrong for ordinary
>>knowledge. Perhaps I should add that it is wrong for AI, and this is
>>why projects such as CYC, based on that bogus concept of knowledge,
>>will never succeed.
>It is hard to imagine ANY interesting intelligence, machine or
>human-like, that did not (a) have representations (models) of itself
>and the world, and (b) have some means of testing the adequacy of
>those representations.
If we take the conclusions of epistemology (so-called) as such a
representation, then that representation fails the adequacy test.
Seth Russell
> What is being looked for is not the mathematical truth, but the
> method used to unearth that truth. The fact that the method produced
> a true statement is important in judging the usefulness of the
> method. But the truth itself is not the mathematical knowledge.
Ok I think i see what you are saying. You point out that knowledge is
the ~process~ of arriving at match points - rather than the match points
themselves. The match points themselves are quite useless without the
process of 'knowing' them. But where X symbolizes a match point
and P symbolizes the process that generates it, we can certainly
distinguish
between X and P and there certainly is a web of 'knowledge' that discusses
both X and P as distinct from each other. You're not suggesting that this
web of knowledge should should only discuss P's and not X's, are you?
Seth Russell
Http://RobustAi.Net/Ai/SymKnow.htm
Neil W Rickert
>Neil W Rickert wrote:
Ps are important because we can use them to generate Xs. Often they
also help to provide the meaning of the Xs. Of course you cannot
omit the Xs. However, many of the Xs are easy to find, while the Ps
are more difficult. Moreover we judge the Ps by their utility, and
the Xs by their truth. My complaint is that traditional epistemology
is unable to distinguish between Ps and Xs.
James Buddenhagen
[...snip...]
> Let me give another example.
>
> that b is a 450 digit prime number. Let's suppose that when you
> multiply them, you get c, which is then a 999 or 1000 digit composite
> number.
>
> If you submitted an article entitled "a * b = c", but with no proof,
> your article would be rejected by any respectable math journal. If
> you submitted it with a proof, and the proof consisted of multiplying
> a and b, and showing that the answer is c, then your article would be
> rejected by any respectable math journal. If, however, you submitted
> it with a proof that started with c, and was cleverly able to come up
> with a factorization without any prior knowledge of a and b, without
> using anything special that makes factoring c unusually easy compared
> to other numbers of a similar magnitude, and without any reliance on
> lucky guesses, then the article would likely be published, and there
> that current the state of the art of factorization is not anywhere
> near being able to deal with 1000 digit numbers that are products of
> two primes.
>
> method used to unearth that truth. The fact that the method produced
> a true statement is important in judging the usefulness of the
> method. But the truth itself is not the mathematical knowledge.
>
All elliptic curves are modular.
Is that not an example of mathematical knowledge?
If I could understand the proof I would have much more knowledge.
That will never happen for me.
One (or two) years ago nobody knew the proof.
At that time the fact that all elliptic curves are modular
was not part of anyone's mathematical knowledge. Now it is.
Now (I feel) it is part of my knowledge.
Mathematics is a collective activity. We cannot all share in
understanding of proofs. But because someone proved that all
elliptic curves are modular, the fact that all elliptic curves
are modular can be knowledge for all.
--Jim Buddenhagen
Seth Russell
http://x23.deja.com/getdoc.xp?AN=557979677 ]
Seth:
But where X symbolizes a match point [fact] and P symbolizes the process
that generates it, we can certainly distinguish between X and P and there
certainly is a web of 'knowledge' that discusses both X and P as distinct
from each other.
Neil W Rickert wrote:
Ps are important because we can use them to generate Xs. Often they also
help to provide the meaning of the Xs. Of course you cannot omit the Xs.
However, many of the Xs are easy to find, while the Ps are more difficult.
Moreover we judge the Ps by their utility, and the Xs by their truth. My
complaint is that traditional epistemology is unable to distinguish between
Ps and Xs.
Seth (new):
Ok, i'd sure buy a P over an X any day and i'm not particularly concerned
with "traditional epistemology" or it's wiggy definition "justified true
belief". I am interested, however, in how P-abilities create X-facts, and
how the X-facts get communicated in our culture and form a tapestry of
knowledge apart from the Ps. It seems to me that tapestry of X-facts
forms the content of our ~conscious~ mind. We are certainly less likely to
be ~aware~ of the processes that originally created the facts and
frequently those processes are not even located within the confines of our
personal brains. It seems to me that if we can trace our facts to the
processes that create them, then our knowledge is on firm ground, if we
cannot, than it is not. The 'process' of X-facts building on other
X-facts ad infinitum (without justification to their P-abilities) is much
of what intelligence appears to me to be. If we knew that process, then we
could certainly do AI. Don't you agree?
Seth Russell
Http://RobustAi.Net/Ai/Conjecture.htm
Neil W Rickert
>[Your response never got to my news server, so i'll reconstruct from
>http://x23.deja.com/getdoc.xp?AN=557979677 ]
I think our main news feed is overloaded, and sometimes it breaks for
a while. The response may show up later.
>Seth:
>But where X symbolizes a match point [fact] and P symbolizes the process
>that generates it, we can certainly distinguish between X and P and there
>certainly is a web of 'knowledge' that discusses both X and P as distinct
>from each other.
>Neil W Rickert wrote:
>Ps are important because we can use them to generate Xs. Often they also
>help to provide the meaning of the Xs. Of course you cannot omit the Xs.
>However, many of the Xs are easy to find, while the Ps are more difficult.
>Moreover we judge the Ps by their utility, and the Xs by their truth. My
>complaint is that traditional epistemology is unable to distinguish between
>Ps and Xs.
>Seth (new):
>Ok, i'd sure buy a P over an X any day and i'm not particularly concerned
>with "traditional epistemology" or it's wiggy definition "justified true
>belief". I am interested, however, in how P-abilities create X-facts, and
>how the X-facts get communicated in our culture and form a tapestry of
>knowledge apart from the Ps.
Rickert (new):
This start where the Ps and Xs were mathematical, although I am more
interested in the case where they are empirical. Let me give some
examples of each.
Rickert (new):
The Pythagorus theorem of geometry might at first look like an X.
However, it is a major prerequisite for the cartesion
coordinatization of space (analytic geometry). In terms of
mathematics, that makes it an important mathematical P.
Rickert (new):
In turn the cartesian coorinatization was one of the stepping stones
to Newtonian physics, and Newton's used of calculus. So that gives us
another mathematical P (the calculus), and an empirical P (the Newtonian
program).
Rickert (new):
To give a more specific example, Ohm's law is an empirical P, for it
establishes the basic principles we need to build voltmeters,
ammeters, Ohmmeters, etc. Each reading of such a meter is an X.
> It seems to me that tapestry of X-facts
>forms the content of our ~conscious~ mind. We are certainly less likely to
>be ~aware~ of the processes that originally created the facts and
>frequently those processes are not even located within the confines of our
>personal brains. It seems to me that if we can trace our facts to the
>processes that create them, then our knowledge is on firm ground, if we
>cannot, than it is not. The 'process' of X-facts building on other
>X-facts ad infinitum (without justification to their P-abilities) is much
>of what intelligence appears to me to be. If we knew that process, then we
>could certainly do AI. Don't you agree?
Rickert (new):
I am suggesting that Ps have more to do with intelligence than Xs.
The brain is using trial and error methods of reorganizing itself,
thus setting up hypothetical Ps. These are then tested for pragmatic
effectiveness (they help to get your basic hunger, thirst and other
drives satisfied). Once effective Ps are found, they are used to
generate Xs, which are the content of perception and thought.
Seth Russell
> I am suggesting that Ps have more to do with intelligence than Xs.
> The brain is using trial and error methods of reorganizing itself,
> thus setting up hypothetical Ps. These are then tested for pragmatic
> effectiveness (they help to get your basic hunger, thirst and other
> drives satisfied). Once effective Ps are found, they are used to
> generate Xs, which are the content of perception and thought.
I have no quarrel with that, but I think it tells half of the story:
P -yield- X. But most of the Xs that our conscious mind comes
in contact with have no associated Ps in our own brain that yield them.
Rather they come to our conscious mind from our culture through
absorption processes (education and research) that have nothing to do
with the process that yielded the original X. These Xs (which don't
come from Ps that we own) stick together in our minds (sorry that's
the best way i can put it) giving us the effect of:
PSickey(X1,X2.X3..). -yield- Xn. Where PStickey is some intelligence
process. Jim Buddenhagen's example is a case in point. Andrew Wiles
had the PStickey for ( X(Taniyama-Shimura conjecture),
X(Kolyvagin-Flach method),....) -yield- X(proof of FLT). The pieces of
external knowledge stuck together in Wiles's head to yield the proof.
I know i've being a bit chiggy here with the quasi mathematical notation,
but my point (that you still have not acknowledged) is that much of
intelligence is based on abilities for processing facts to yield other facts,
without needing to know (or caring) how the facts are grounded.
Seth Russell
Neil W Rickert
>Neil W Rickert wrote:
>> I am suggesting that Ps have more to do with intelligence than Xs.
>> The brain is using trial and error methods of reorganizing itself,
>> thus setting up hypothetical Ps. These are then tested for pragmatic
>> effectiveness (they help to get your basic hunger, thirst and other
>> drives satisfied). Once effective Ps are found, they are used to
>> generate Xs, which are the content of perception and thought.
>I have no quarrel with that, but I think it tells half of the story:
>P -yield- X. But most of the Xs that our conscious mind comes
>in contact with have no associated Ps in our own brain that yield them.
I don't agree.
>Rather they come to our conscious mind from our culture through
>absorption processes (education and research) that have nothing to do
>with the process that yielded the original X.
It surely took something to convert those vibrating air molecules
into an X in my brain. Why don't you consider that involves Ps?
> These Xs (which don't
>come from Ps that we own) stick together in our minds (sorry that's
>the best way i can put it) giving us the effect of:
You measure your window. To do that you use Ps, including the ruler
and all of the procedures that you follow in using the ruler. Now
you send that length to me; it is an X. As it happens, I have often
used similar Ps to measure things. Therefore I am in a good position
to appreciate that X. But if I had never used a ruler in my life,
and had no idea what measuring with a ruler entailed, then I would
not have the similar Ps, and I also wouldn't find anything meaningful
in the X that you sent me.
>PSickey(X1,X2.X3..). -yield- Xn. Where PStickey is some intelligence
>process. Jim Buddenhagen's example is a case in point. Andrew Wiles
>had the PStickey for ( X(Taniyama-Shimura conjecture),
>X(Kolyvagin-Flach method),....) -yield- X(proof of FLT). The pieces of
>external knowledge stuck together in Wiles's head to yield the proof.
Wiles used lots of P's.
>I know i've being a bit chiggy here with the quasi mathematical notation,
>but my point (that you still have not acknowledged) is that much of
>intelligence is based on abilities for processing facts to yield other facts,
>without needing to know (or caring) how the facts are grounded.
I won't acknowledge what I do not believe. A sentence for which you
do not know the meaning (i.e., which is not grounded) is not any kind
of fact. It is just a sequence of squiggly marks.
kenneth Collins
> Seth Russell <se...@halcyon.com> writes:
>[...]
> >I know i've being a bit chiggy here with the quasi mathematical notation,
> >but my point (that you still have not acknowledged) is that much of
> >intelligence is based on abilities for processing facts to yield other facts,
> >without needing to know (or caring) how the facts are grounded.
>
> I won't acknowledge what I do not believe. A sentence for which you
> do not know the meaning (i.e., which is not grounded) is not any kind
> of fact. It is just a sequence of squiggly marks.
what Seth says is True, and, if you limit yourself to what you say, than you can
say nothing until you know everything, and i observe that no one knows everything
yet.
there're well-defined neural processes that make short-work of the 'difficulty'
inherent, which include the manifestation of all symbolic representation,
including Maths.
one can represent these neural processes symbolically, but the neural processes
remain the neural processes, which are 'just' energy-flow dynamics.
Maths, like all symbolic manifestations, is (are) 'only' extremely high-'level'
language(s).
if i interpret his comments correctly(?), Seth is viewing the low-'level',
energy-flow, stuff.
and it makes no difference whether or not you 'acknowledge' this stuff. it stands
Proven that it's in-there.
K. P. Collins
Seth Russell
> >I have no quarrel with that, but I think it tells half of the story:
> >P -yield- X. But most of the Xs that our conscious mind comes
> >in contact with have no associated Ps in our own brain that yield them.
>
> I don't agree.
Ok, *all* of the Xs have Ps associated, otherwise they would not even reach our
consciousness. I guess I didn't fully appreciate that when we started this
dialogue.
> >Rather they come to our conscious mind from our culture through
> >absorption processes (education and research) that have nothing to do
> >with the process that yielded the original X.
>
> It surely took something to convert those vibrating air molecules
> into an X in my brain. Why don't you consider that involves Ps?
Yes, the "absorption processes" (mentioned above) are very definitely Ps. My
point is only that they are not necessarily the same Ps that created the Xs. They
may not even be related. Example: there is a giant industry (P tool) devoted to
flying people through the air. Part of the function of that industry is to tell me
the time of departure of my airplane. In this case the network of tools (external
and internal to my brain) by which I become aware of the departure time of my
flight has nothing to do with the network of tools that causes the airplane fly
off the ground at that time.
> You measure your window. To do that you use Ps, including the ruler
> and all of the procedures that you follow in using the ruler. Now
> you send that length to me; it is an X. As it happens, I have often
> used similar Ps to measure things. Therefore I am in a good position
> to appreciate that X. But if I had never used a ruler in my life,
> and had no idea what measuring with a ruler entailed, then I would
> not have the similar Ps, and I also wouldn't find anything meaningful
> in the X that you sent me.
The X could mean something entirely different to you than it means to me and still
be the *same* X, and still be just as grounded in your Ps as it is in my Ps. Xs
become independent of Ps.
> I won't acknowledge what I do not believe. A sentence for which you
> do not know the meaning (i.e., which is not grounded) is not any kind
> of fact. It is just a sequence of squiggly marks.
Yes I agree.
But if
Mind1 [P1 --yield- X1]
X1 --communicated-- Mind2
Mind2 [P2(X1) --yield-- X2)
Can we say anything meaningful about the the similarity of P1 to P2 ?
Seth Russell
Patrik Bagge
My
>point is only that they are not necessarily the same Ps that created the
Xs. They
>may not even be related. Example: there is a giant industry (P tool)
devoted to
>flying people through the air. Part of the function of that industry is to
tell me
>the time of departure of my airplane. In this case the network of tools
(external
>and internal to my brain) by which I become aware of the departure time of
my
>flight has nothing to do with the network of tools that causes the airplane
fly
>off the ground at that time.
interesting example. It seems that we empirically connect the X1
departure timetable with the real event X2. if X2 is accurate enough
to X1, we tend to equal X1,X2. But when X2 shows variations to X1
then we first get 'annoyed' and maybe begin to investigate P2
(processes behind departure) and even create Xs
(reasons/factors behind late departures) and maybe even Xs
(ways of avoiding to be late to the meeting)
example, in bad weather, book an earlier flight.
To me it seems that Xs are actually Ps in our nervous system until
we attempt to communicate them, then they become a X.
>But if
>Mind1 [P1 --yield- X1]
>X1 --communicated-- Mind2
>Mind2 [P2(X1) --yield-- X2)
>Can we say anything meaningful about the the similarity of P1 to P2 ?
i'd say it depends on the current state or similarities in obtained
knowledge
between Mind1, Mind2. If we assume that the communication is successfull
(same language), then a communicated symbol of 'red' assumes similar
visionary perception (Pn). The difference between P1,P2 becomes quite
obvious
when the spectrum is slowly tuned towards orange. The
perception/communicated
symbol is likely not to occur at the same spectral wavelength for Mind1,2.
Such is the 'inaccuracy' or subjectivity of our vision/nervous system
+ obtained knowledge.
Regards
Patrik Bagge
Neil W Rickert
>Neil W Rickert wrote:
>> >I have no quarrel with that, but I think it tells half of the story:
>> >P -yield- X. But most of the Xs that our conscious mind comes
>> >in contact with have no associated Ps in our own brain that yield them.
>> I don't agree.
>Ok, *all* of the Xs have Ps associated, otherwise they would not even reach our
>consciousness. I guess I didn't fully appreciate that when we started this
>dialogue.
Right.
>> >Rather they come to our conscious mind from our culture through
>> >absorption processes (education and research) that have nothing to do
>> >with the process that yielded the original X.
>> It surely took something to convert those vibrating air molecules
>> into an X in my brain. Why don't you consider that involves Ps?
>Yes, the "absorption processes" (mentioned above) are very definitely Ps. My
>point is only that they are not necessarily the same Ps that created the Xs. They
>may not even be related.
I agree with that.
>The X could mean something entirely different to you than it means to me and still
>be the *same* X, and still be just as grounded in your Ps as it is in my Ps. Xs
>become independent of Ps.
That's possible too.
>> I won't acknowledge what I do not believe. A sentence for which you
>> do not know the meaning (i.e., which is not grounded) is not any kind
>> of fact. It is just a sequence of squiggly marks.
>Yes I agree.
>But if
>Mind1 [P1 --yield- X1]
>X1 --communicated-- Mind2
>Mind2 [P2(X1) --yield-- X2)
>Can we say anything meaningful about the the similarity of P1 to P2 ?
That is somewhat doubtful. One might say that there is a similarity
in what they do, but very different things can perform similar
functions.
Seth Russell
> To me it seems that Xs are actually Ps in our nervous system until
> we attempt to communicate them, then they become a X.
Well for them to be communicated, they certainly must be Xs. But I think that
making a binary distinction between Ps and Xs (both inside and outside of a
mind) will be like separating the top layer of a can of worms. It might be
interesting to look at a different duality: being and sign. Here we can always
make the distinction by using the 'arbitrary substitution test' to determine if
it is a sign.
Seth Russell
Http://RobustAi.Net/Ai/SymKnow.Htm
Patrik Bagge
>> we attempt to communicate them, then they become a X.
>
>Well for them to be communicated, they certainly must be Xs. But I think
that
>making a binary distinction between Ps and Xs (both inside and outside of a
>mind) will be like separating the top layer of a can of worms.
yes, this can of worms is troublesome, when referring to the nervous system,
but i'm beginning to get the 'feel' for it. Ken has some good chapters in
aok
and certainly my 5 1/2 year old is providing me with precious knowledge
regarding the reallife development of a mind's capabilities.
(skills & abstraction levels)
So far i haven't found any unexplanatory capabilities, they all seem to
be grounded in reallife objects and repetetive chain-of-events.
One more comment about the Xs, it seems that the strength or
subjective validity of a X is based on repetiton and an external
majority vote. The nervous system makes up it's own subjective
x that is most of the time close to the objective X.
In science it's often done by empirical verification.
>It might be
>interesting to look at a different duality: being and sign. Here we can
always
>make the distinction by using the 'arbitrary substitution test' to
determine if
>it is a sign.
i'm not familiar with that test, what's it about?
>Seth Russell
>Http://RobustAi.Net/Ai/SymKnow.Htm
That's one URL that i like, have you decided upon any guidelines
for the naturally occurring robust ai ?
Yours
Patrik Bagge
Seth Russell
> yes, this can of worms is troublesome, when referring to the nervous system,
> but i'm beginning to get the 'feel' for it. Ken has some good chapters in
> aok
Aok was the first time I've been exposed to a theory of "How the brain works"
from a 'energy/entropy' perspective and as such it's intriguing. But, KPC is
right, you need a good text on nuroanatomy to understand it, and Ken himself
seems unwilling to explain it to the likes of me. So, since I can't understand
it; and since it seems to conclude things like 'There is an absolute Truth which
humans can know' (which i can't swallow); I guess it will need to wait for a
more articulate spokesman before it enters my consciousness.
> One more comment about the Xs, it seems that the strength or
> subjective validity of a X is based on repetiton and an external
> majority vote. The nervous system makes up it's own subjective
> x that is most of the time close to the objective X.
> In science it's often done by empirical verification.
I tend to agree. But I think there's something that happens to Xs (information)
when they become independent of their Ps (processes) - they spin around and form
a tapestry, moving and interacting with no regard for their grounding - i.e.
they form Gestalts. We can see this happening in the mind internal to the brain
by introspection. Someday, perhaps, there will be scientific tools to observe
that happening in 'mind' external to the brain (of course, first, we need to
believe that mind exists in culture outside of our brains).
> >It might be interesting to look at a different duality: being and sign.
> >Here we can always make the distinction by using the
> > 'arbitrary substitution test' to determine if it is a sign.
>
> i'm not familiar with that test, what's it about?
If any arbitrary value is substituted for each occurrence of a sign
(reference/pointer in a mind), the sign retains its exact functioning to the
subjective mind. You cannot do that with processes that are not signs.
> >Http://RobustAi.Net/Ai/SymKnow.Htm
>
> That's one URL that i like, have you decided upon any guidelines
> for the naturally occurring robust ai ?
I'm more in a 'find it' mode, than a 'invent' mode. The closest thing I've
found is the new Semantic Web that is starting to happen around RDF. It's very
close to SVO at the URL above. The thing is they're working on the external
mind ... someone needs to start making our tools for the internal (subjective)
minds that will interface with this new Web.
Seth Russell
Http://RobustAi.Net/Ai/Conjecture.htm
Patrik Bagge
>> yes, this can of worms is troublesome, when referring to the nervous
system,
>> but i'm beginning to get the 'feel' for it. Ken has some good chapters in
>> aok
>
>Aok was the first time I've been exposed to a theory of "How the brain
works"
>from a 'energy/entropy' perspective and as such it's intriguing. But, KPC
is
>right, you need a good text on nuroanatomy to understand it, and Ken
himself
>seems unwilling to explain it to the likes of me.
well, the reason(s) for this behaviour seems quite obvious...
> So, since I can't understand
>it; and since it seems to conclude things like 'There is an absolute Truth
which
>humans can know' (which i can't swallow); I guess it will need to wait for
a
>more articulate spokesman before it enters my consciousness.
yes, some parts are somewhat difficult, i wish people could understand
and accept that we all are not hot neuro-scientist, thus our expertice might
exist elsewhere.The 'absolute truth' term could be interpreted in many ways
, one truth might be the knowledge of the workings of the nervous system
information processing. Once we have understood this truth, the subjective
nature of our perception might become a truth and furthermore our emitted
behaviour might become clearer. Understanding the factors behind our
behaviour, will probably change our behaviour.
Anyway, there are parts that i like and of course, interpret the way i wish.
The text regarding the predators automatic avoidance of objects moving
towards our vision or even reaction to certain high volume/frequency sounds,
strongly supports 'my' pre-programmed survive&(reproduce) basic instinct.
We don't have to learn it, it's 'in there'
>> One more comment about the Xs, it seems that the strength or
>> subjective validity of a X is based on repetiton and an external
>> majority vote. The nervous system makes up it's own subjective
>> x that is most of the time close to the objective X.
>> In science it's often done by empirical verification.
>
>I tend to agree. But I think there's something that happens to Xs
(information)
>when they become independent of their Ps (processes) - they spin around and
form
>a tapestry, moving and interacting with no regard for their grounding -
i.e.
>they form Gestalts. We can see this happening in the mind internal to the
brain
>by introspection. Someday, perhaps, there will be scientific tools to
observe
>that happening in 'mind' external to the brain (of course, first, we need
to
>believe that mind exists in culture outside of our brains).
EEG etc, techniques are getting quite fancy and probably will improve a lot
in the future. It's a tough problem, like picking out a favourite tune
amongst
a million radio-transmitters, that continually change carrier frequency.
When you say cultural mind, do you mean the 'sum' of many individual
minds as a whole?. A society could certainly be looked upon this way,
since we all communicate with each other. It seems that the 'physics'
behing marketing adresses the issue of programming the society mind.
anyway, the links&images ideas, some time ago, i do beleive captures
the essence of the dynamics behind Ps and Xs, I'm continuing to work
on it and hope to refine it into an application.
>> >It might be interesting to look at a different duality: being and sign.
>> >Here we can always make the distinction by using the
>> > 'arbitrary substitution test' to determine if it is a sign.
>>
>> i'm not familiar with that test, what's it about?
>
>If any arbitrary value is substituted for each occurrence of a sign
>(reference/pointer in a mind), the sign retains its exact functioning to
the
>subjective mind. You cannot do that with processes that are not signs.
could you give an example?
>> >Http://RobustAi.Net/Ai/SymKnow.Htm
>>
>> That's one URL that i like, have you decided upon any guidelines
>> for the naturally occurring robust ai ?
>
>I'm more in a 'find it' mode, than a 'invent' mode. The closest thing I've
>found is the new Semantic Web that is starting to happen around RDF. It's
very
>close to SVO at the URL above. The thing is they're working on the
external
>mind ... someone needs to start making our tools for the internal
(subjective)
>minds that will interface with this new Web.
ok, i'm working on a private site, currently trying to learn some more java,
in order
to create public demos. This seems to be a good way of 'marketing' one's
ideas.
There are a lot of nice java demos out there, often ALife applications.
The dynamics in desire(goal) / behaviour / loopback(feeling) is interesting,
i have done some simulations with a limited set of these. When one gets
the constants and interaction between feelings to a reasonable level, it
looks almost 'human like'
example: survive->hunger minimizes the survive->curiosity when time
elapses and emitted behaviour does not 'find food', thus directing more
attention to 'find food' behaviour etc.
Yours
Patrik Bagge
PS, let's 'hang around' 30 years or so and see what kind of robust AI
cake we can bake ... there is no hurry.
Patrik Bagge
>I have also read Ken's manifesto, and I find it impressively thorough.
>However it is possible to get lost "amongst all the trees" that way.
>To me, it seems that his manifesto does not straight-forwardly address the
>key underlying problem we are up against when we try to understand
>ourselves.
hmm, yes, iv'e spend hours with the text, many neuro terms make
it rather difficult to get 'thru', but i beleive that i have captured some
of the 'smell' of it. i'd be glad to hear your interpretation of the key
elements in the text.
>His approach partly avoids truths that tend towards becoming distressfully
>associated by having kept the manifesto's main explanatory focus on
>neurophysical "technicalities", and partly also because he does not make
use
>of and thoroughly enough (or at all) integrate (or explicitely enough take
>into account) the incredibly powerful *scientifically established*
>principles and (Darwinian) theory of our evolution.
yes, my impression also, iv'e 'bugged' the man a bit with 'my'
survive&reproduce pre-programmed human primary directive.
First he said B.S, later reversing back a bit on it.
I'm not clear at all where he 'stands' in this issue.
>It is to me indicative, that Ken has not understood (or it seems not even
>bothered to try to understand) my complementary and likewise faithfully
>science-aligned (as far as I know) explanatory scheme -- a scheme that does
>make use of both fundamental physics (albeit most philosophically and
>peripherally), Evolution Theory, and relevant results and interpretaions
>from fields of brain and behavioural science. (He has not manged to level
>even one piece of specified criticism of my scheme.)
>
>Yours ever only partly decEPTively joking,
>
>Peter F. (usually hiding under the alias "Sir Knowitall")
aah, it's you, well i'm following yours & kens conversation
with my amatourish perception in the neuro area.
>http://web.one.net.au/~fellin/main.htm
i will B-mark and read thru... hopefully you
will bear some questions later on ..
>P.S. (Are you, Patrik Bagge also a Swede?
jadå, absolut, du med?
jobbar i norge just nu, men ær vældigt stockholmsk
Åker øver till jul å nyår, verkar vara bra partaj i stan
(100 mil. nedlagt ! )
In english, Yes.
Yours
Patrik Bagge
Peter Fellin
However it is possible to get lost "amongst all the trees" that way.
To me, it seems that his manifesto does not straight-forwardly address the
key underlying problem we are up against when we try to understand
ourselves.
His approach partly avoids truths that tend towards becoming distressfully
associated by having kept the manifesto's main explanatory focus on
neurophysical "technicalities", and partly also because he does not make use
of and thoroughly enough (or at all) integrate (or explicitely enough take
into account) the incredibly powerful *scientifically established*
principles and (Darwinian) theory of our evolution.
It is to me indicative, that Ken has not understood (or it seems not even
bothered to try to understand) my complementary and likewise faithfully
science-aligned (as far as I know) explanatory scheme -- a scheme that does
make use of both fundamental physics (albeit most philosophically and
peripherally), Evolution Theory, and relevant results and interpretaions
from fields of brain and behavioural science. (He has not manged to level
even one piece of specified criticism of my scheme.)
Yours ever only partly decEPTively joking,
Peter F. (usually hiding under the alias "Sir Knowitall")
http://web.one.net.au/~fellin/main.htm
P.S. (Are you, Patrik Bagge also a Swede?
-----
Patrik Bagge <p...@neramd.no> wrote in message
news:ub264.2597$Sfm.18...@news.telia.no...