X-Message-Number: 8569
Date: Tue, 09 Sep 1997 10:39:34 -0400
From: "John P. Pietrzak" <>
Subject: Re: Digital Shakespeare
References: <>

John K Clark wrote:
> To repeat, the task I gave you will terminate and is not impossible,
> I thinks it's difficult but prove me wrong and tell me the trillionth
> digit of PI.

Well, what has it been, four or five messages now?  What can I say
that I haven't said before?  I've already said that it (probably)
takes more time than I have available to me in my entire life, so I'm
not goint to give you the digit itself.  I've already said that I still
don't think that makes it difficult or complex, just time consuming.
I've given you a truly trivial algorithm, something practically any
child could execute.  What more do you want!!!!

>         >And I do state, the process of following the steps in the
>         >algorithm is simple, something most every human on the
>         >planet can do.
> 
> You find it simple so do it and tell me the results, I tried to do it
> myself but found it too complex, I guess that makes me a simpleton.

Ridiculous!  Which of these operations did you find too complex:
Addition?  Subtraction?  Division?  Variable assignment?  There isn't
anything else to the algorithm.

You're just saying that time equals complexity here.

[On words]
{a}
>         >>I have no definition of "simplicity" or "complexity" just
>         >>as I have none for "intelligence" and for the same reason,
>         >>these things are best explained by example.
> 
>       >[JP remarks about training people with those examples]
{b}
> Actually when I talk to people I usually use words,
{c}
> but I've looked up very few of those words in a dictionary because I
> didn't need too, and even when I did it doesn't weaken my case
> because the lexicographers got the knowledge to write their book the
> same way I did, from usage.
>
> By itself a dictionary is nothing but one big circular definition.
> All the definitions in a dictionary are made of words, and those
> words also have definitions made of other words also in the
> dictionary, and round and round we go.[...]
{d}
> In the real Physical world dictionaries are useful because there is a
> lot of information that all readers have in common that is not in any
> book.
{e} 
> If I am frowning and have tears in my eyes and say "I am sad" then
> you don't need any book to have some idea of what I'm feeling. If you
> then read that "happy is the opposite of sad" then you understand in
> a deep way what "happy" means, it is no longer just a squiggle, it is
> a symbol with meaning.

Ok, let me see if I've got this straight.  In the first quoted portion
{a}, you admit to a purely connectionist frame of meaning, words are
merely the titles to exemplar-based categories.  However, in part {b},
you state that you use words, not examples, to communicate with others,
even (presumably) those you have never met before and couldn't possibly
have seen the same examples you have.  Now, in {c}, you return to saying
that words by themselves are meaningless.  However, you modify this in
{d} by saying that, somehow, simple lists of words aren't completely
meaningless, in fact they are generally useful, based upon the idea
that "all people have information in common" (so everyone has seen the
same examples?) and can therefore remove the circularity in the
definitions.  Finally, in {e}, you emphasize this point by saying
that people can do *much more* than just categorize, they can find the
ontological complement of a category just by thinking about it.  (I.E.,
people can reason.)  (Plus, you seem to think that I've already built
up a category around tears and frowns...)

My question, then, is: are you arguing for connectionism here or not?
How can all people, requiring examples to train their neural nets, have
a base of information in common?  How can a purely connectionist person
find the opposite of a concept, when reasoning requires rules, axioms,
definitions?

[On Shakespeare vs. Random]
> You don't have to convince me that Shakespeare is greater than
> gibberish and even in some way more complex, what I'm saying is that
> I don't have a mathematical definition of "complex" to go with my
> intuitive understanding of the word and you don't either. I wish you
> did.

I wasn't trying to make a statement about complexity there.  Shakespeare
consistently produced text which was a very constrained subset of the
total possible random sequences, and those constraints moreover followed
a significant pattern: the text followed the rules of the English
language, and provided some tangible meaning to humans which read the
text.  (In other words, not only was it not gibberish, it wasn't
properly formatted English gibberish, it actually has semantic content
for readers.)

Therefore, Shakespeare ASCII is significantly different from the vast
majority of normal random ASCII.

[On the intelligence of Einstein]
> I wasn't born yesterday, I just don't believe that when you think of
> Einstein you first think of a man who was able to feed himself.

Unfortunately, you didn't ask me for my first thought, you asked me
whether I thought he was intelligent.  No, sir, I DON'T associate
intelligence with mathematics or physics.  Perhaps only intelligent
people become mathematicians or physicists, but that doesn't mean that
only mathematicians or physicists are intelligent people.  I'm looking
for a more general definition of intelligence than a person's
occupation.

> Perhaps I should have used Steven Hawking in my example because he
> can do none of the things your list.

Steven Hawking has written several best-selling books.  Steven Hawking
has taken part in a number of TV documentaries.  Steven Hawking has
made an appearance on Star Trek: The Next Generation.  He may not be
able to move many of the muscles in his body, but he can do quite a
number of the things on my "list", many of which are not associated
with physics or mathematics.

[More on the definition of intelligence]
>         >When I'm trying to be precise, however, I need to be more
>         >careful about how I use the term, particularly when I'm
>         >trying to use it in an objective manner.
> 
> In other words one set of ideas is useful in everyday life and
> another set is useful when arguing philosophy on Cryonet.

Ah, Grasshopper, you begin to see the truth.  Only a slight modification
of your statement above, and you will see all.  "In other words, one
set of ideas is useful in everyday life and another set is useful when
arguing philosophy."

[On the Turing Test]
> Obviously anybody can be a tester and equally obviously that will
> change the results of the test, just as in real life. I thought we
> already both agreed that The Turing Test was as subjective as a test
> that determined what was interesting and what was not.

I thought we had, too, but you were criticizing Eliza's passing the
Turing Test based upon the _tester_, not the test.  If it doesn't
matter who the tester is, why do you make such a fuss about Eliza?
It passed the test, fair and square.


John

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=8569