Re: Crick's answer to lester's conundrum

On 27 Apr 2006 09:50:58 -0700, "feedbackdroids"
<feedbackdroids@xxxxxxxxx> in comp.ai.philosophy wrote:


Lester Zick wrote:


BTW, whether you know it of not, the idea of D-o-D has been around for
at least ... 20 years ....

Oh do tell at least twenty years? Try twenty-five years.

pg 238 of Society of Mind, 1986, section 23.1 ....

"... not only are differences important by themselves; more often than
we realize, we think about 'differences between differences' ...

Sure especially since my own copyrighted work on the subject dates
to 1982. I don't doubt it's been plagiarized since. In other words
according to you I should be "thrilled to see others citing my work"
and calling it their work.



Now, I begin to see - as I had suspected all along - what D-o-D really
is. One in a bag of tricks used by intelligent systems. What "is" the
Society of Mind? Namely, a large bag of tricks used by intelligent
systems.

We appreciate your opinion on this subject just as we appreciate all
your opinions on a variety of subjects.

BTW, Lester, I see your 25 years, and raise you 35 .......
D[D[A:B]:D[C:X1]] .... do the math.
We see that the quote in SOM is just a non-mathematical way of saying
the same thing as the following from 1971.

http://web.media.mit.edu/~minsky/papers/PR1971.html

===============
Sec 4.6 Analog, again

...............

In particular, two difference-descriptions can be compared as handily
as any other pair of descriptions.

"Compared" how exactly? "Paired" how exactly?

Now we can apply this idea to the analogy problem. The machine must
select that scene X (from a small collection of alternatives) which
best completes the statement

"Select" "scene" "small" "collection" "alternatives" "completes" ????

My how the list of undefined terminology doth expland apace to paper
over the mechanical holes in your never ending assumptions.

A is to B as C is to X

"Is" "to" "as" ???? My you're just full of unexplained and
mechanically unreduced prepositions and conjunctions.

My answer to this is simply AI is to Science as KP is to the Army and
those with no discernible professional talent will eventually wind up
in one or the other.

That is, one must find how B relates to A and find an X that relates to
C in the same way. Using the expression D[A:B] to denote the
difference-description-network resulting from comparing A with B, we
simply compare the structures resulting from:

D[D[A:B]:D[C:X1]],
D[D[A:B]:D[C:X2]],
D[D[A:B]:D[C:X3]], etc.

Each of these summarizes the discrepancies within the "analogical
explanations" for each corresponding possible answer. So to make the
decision, we have to choose the "best" or "simplest" of these. We will
not give details of how this is done; it is described in Chapter 7 of
Winston's thesis.

"Summarizes" "discrepancies" "analogical" "explanations" "each"
"corresponding" "possible" "answer" "decision" "choose" "simplest"
"best" ???? And the list of undefined and unregressed terminology just
goes on and on. Perhaps you'd just care to pull another rabbit out the
the old philosophical grabbag? Or when you do get around to doing a
little mechanics for a change please advise.

~v~~

.