Re: [ontolog-forum] Semantics of Natural Languages

["Rich Cooper" <rich@xxxxxxxxxxxxxxxxxxxxxx>]

Hi Azamat, Adrian and John,

Comments below,

-Rich

Sincerely,

Rich Cooper

EnglishLogicKernel.com

Rich AT EnglishLogicKernel DOT com

9 4 9 \ 5 2 5 - 5 7 1 2

-----Original Message-----
From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx [mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of AzamatAbdoullaev
Sent: Wednesday, August 31, 2011 9:17 AM
To: [ontolog-forum]
Subject: Re: [ontolog-forum] Semantics of Natural Languages

AA:>On Wednesday, August 31, 2011 6:50 AM, John wrote:

"Cyc is probably the ultimate development of the classical AI paradigm of a

big ontology + knowledge base + inference engine(s).

About a dozen years ago, I was talking with Mike Genesereth, who said "Lenat

is probably the only one who doesn't know that Cyc has failed."

RC:> That is the kind of thinking that all of us show in one form or another.  We seem stuck in our structured ways after the first four decades or six, unable to overthrow the past beliefs and institute new untried ones. 

This is an interesting phenomenon, meriting another special ontology, of

critical thinking, as a high-order thinking questioning assumptions and the

quality of methods, historically originated as Socratic method. Its about

willingness to critically evaluate your thinking, remain open to alternative

perspectives, to integrate new/revised perspectives into your ways of

thinking and acting, employing ontology and logic, formal and informal,

intelectual skills, as breadth and depth, credibility and relevance,

clarity, accuracy and precision, significance and fair-mindedeness, as well

as intellectual traits: integrity, tenacity, confidence, courage, empathy,

and i-humility .

If one's critical thinking, which is still not a creative thinking, is

devoid of the intellectual habits/traits of the mind, then the same subject

falls prey to his egoistic self-interests, he is manupulative, unethical,

and badly subjective.

There is a bad public side here as well, the cause why many academic and

semi-academic projects, well-promoted and well-funded by the government

agencies, are never delivering anything valuable for the public.

Azamat Abdoullaev

RC:> I'm not familiar with Cyc, other than with the original papers Lenat wrote about Eurisko, AM, etc that I assume are precursors of Cyc, with lots of humans editing the materials to make them fit "standard beliefs and nomenclature".  But I haven't heard anything about Cyc being able to do creative thinking beyond what was described in those precursors - combining older simpler concepts into, perhaps novel, newer concepts certainly, but also throwing out some of the old concepts that get in the way of new ones.  Most of creative thinking is in such rearrangement or recombination of old ideas, along with jettisoning past strongly held beliefs, IMHO. 

That is visible in patent specification sequences, especially when the inventor documents the older technologies that led to the one she is patenting; then the next one jettison's hers, and so on.  There are always limitations to current technology, and when "current" gives way to "later", there are parallel developments which only practitioners of the art tend to follow with reasoning instead of duplication. 

I don't see that concept of "being influenced and overthrown by later developments" discussed in Cyc, but then I don't see much of anything discussed about Cyc past the precursors I mentioned anywhere in the public literature; I'm not referring to tutorials about Cyc, but about analyses.  Does anyone have references to how Cyc has progressed since then - publicly available references freely downloadable? 

----- Original Message -----

From: "John F. Sowa" <sowa@xxxxxxxxxxx>

To: <ontolog-forum@xxxxxxxxxxxxxxxx>

Sent: Wednesday, August 31, 2011 6:50 AM

Subject: Re: [ontolog-forum] Semantics of Natural Languages




> Adrian,

>

> I agree with your comments:

>

>> I think John McCarthy once said words to the effect that what's needed

>> for the really big problems in AI (such as general computational NL

>> understanding), is to stop and _think_ for a very long time.

>

> Good advice.  But every once and a while, it's useful to document

> your thinking, otherwise none of that thought ever gets recorded.

> But it's also important to recognize that no single publication

> is ever definitive.  Each document is just one step along the way.

>

RC:> Yes, that is what I meant above.  But even more; the older must be thrown out with the bath water at times to reach new ones.  Ether as the conductor of light, for example. 

>> It's very tempting to hope that something in the philosophical or AI

>> literature, combined with the data structure du jour, plus the current

>> fashion for AI-style ontologies can somehow be glued together to 'solve'

>> general free NL understanding.

>

> I agree.  Cyc is probably the ultimate development of the classical AI

> paradigm of a big ontology + knowledge base + inference engine(s).

> About a dozen years ago, I was talking with Mike Genesereth, who said

> "Lenat is probably the only one who doesn't know that Cyc has failed."

Has anyone offered an explanation why Cyc is judged to have "failed"?  It is, after all, a managed discovery engine managed by a bunch of philosopher kings, from what I read, indoctrinated in the latest thinking from a handful of universities with large endowments.  If philosophy "works" in that sense, there should be at least a focus on the thinking evolving from that handful. 

> I wouldn't say that Cyc is a total failure.  I think that it has some

> useful components.  But it is never going to achieve its original goals

> if it continues along its current trajectory.

RC:> What other trajectory would likely work better to advance Cyc, if any is known or suggested?

>> However, the evidence points the other way: Watson being beaten by

>> Congressman Holt at Jeopardy [1] is only the latest in a painfully long

>> list. (Jeopardy competence is much less than full NL understanding

>> anyway)

>

> Jeopardy is less tightly structured than Cyc, and it therefore has a

> better chance to evolve into something useful. But with more thinking,

> it could have achieved much better results with much less hardware.

RC:> What kind of thinking, about what kinds of things, would have improved Watson at Jeopardy?  Why is Jeopardy relevant other than being a public gladiatorial spectacle?  There is no room for creative thought in Jeopardy, only for matching the hints against historical instances. 

>> So, my money is on someone unexpected, like Einstein in his patent

>> office days, with lots of time to think, and not overly wedded to

>> current paradigms.  Not rushing to implement.  As they used to say at

>> IBM -- _Think_.

RC:> IBM Think led to Microsoft's purchase of DOS and exploitation thereof.  All thinking is extension of previous thinking (except in the newborn), so can't overcome old errors that get encoded into the thought DNA of the structure doing the thinking.  That is why evolution occurs, eliminating the old versions that don’t work well in the new environments.  All new contributions to knowledge seem to have come from outside the old social structures.  Newton in his study, Darwin in his, Einstein motivated by Michelson-Morley(Sp?) experiment, Heisenberg's overthrow of certainty, which has still not been heeded properly, Gödel's proof of algebra's insufficiency, Shockley's rethinking of the overbiased diode, and so many more inventors never publicized. 

> I agree.  The following slides, which I presented at ICCS 2011 in July,

> survey 5 different cognitive architectures, of which Cyc is the first:

>

>    http://www.jfsowa.com/talks/ca4cs.pdf

>    Cognitive architectures for conceptual structures

>

> Each of the architectures is a moving target that has been evolving

> for 20 years or more.  The fifth is the CS architecture, which I

> presented in my 1984 book, _Conceptual Structures_.  Those slides

> discuss the original 1984 version, the developments since then,

> and the relationships to neuroscience.

>

> John

>

_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/  
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/  
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/ 
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J    (01)