ontolog-forum
[Top] [All Lists]

Re: [ontolog-forum] Ontology based conversational interfaces

To: "'Thomas Johnston'" <tmj44p@xxxxxxx>, "'[ontolog-forum] '" <ontolog-forum@xxxxxxxxxxxxxxxx>
From: "Rich Cooper" <metasemantics@xxxxxxxxxxxxxxxxxxxxxx>
Date: Sat, 11 Jul 2015 13:53:40 -0700
Message-id: <030b01d0bc1b$abeae020$03c0a060$@com>

Dear Tom,

 

Thanks for the reference to the Winograd et al work Understanding Computers and Cognition.  I managed to find this 60 page pdf review of it here:

 

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.32.3411&rep=rep1&type=pdf

 

Amazon doesn't offer a kindle edition, and the paperback costs $26 which is discouraging, but if I see it in a bookstore, I will definitely give it a overview read. 

 

Sincerely,

Rich Cooper,

Rich Cooper,

 

Chief Technology Officer,

MetaSemantics Corporation

MetaSemantics AT EnglishLogicKernel DOT com

( 9 4 9 ) 5 2 5-5 7 1 2

http://www.EnglishLogicKernel.com

 

From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx [mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of Thomas Johnston
Sent: Saturday, July 11, 2015 11:17 AM
To: [ontolog-forum]
Subject: Re: [ontolog-forum] Ontology based conversational interfaces

 

Rich,

 

<<< 

So it would seem that knowing these details, one could write a requirements doc about how the conversational interface would help people deal with these tendencies. 

>>> 

 

If a lot of work has been done in a field, it's best not to work out your own theories without knowing something about that work. Watching a talk by one guy from Duke, and then talking about writing a specification, is like undergraduates in a Philosophy class who read selections from Descartes' Meditations, and then write term papers in which they solve the mind-body problem.

 

On the topic of a "conversational interface", I recommend:

 

Terry Winograd and Fernando Flores. Understanding Computers and Cognition, especially Chapter 5, "Language, Listening and Commitment", which takes a speech act approach to understanding how agreement can be reached in conversations.

 

There is also a great deal of relevant work on Paul Grice's rules of conversational implicature.

 

Also there is a lot of recent work on the topic of discourse representation theory. What should be interesting about that work, to ontology engineers, is that it is about creating a formal logic representation of the process by which participants in a conversation begin with a set of assumptions, discover that some of them are not shared by others and that some of them are, discuss the ones that are not shared, and so on and so on. As time passes in the conversation, a set of shared assumptions builds up. At the conclusion of the conversation, the shared assumptions worked out and agreed to can be called a "mini-theory" that the participants agree on. And its conclusions are expressed in a formal framework, something that ontology engineers could go to work on.

 

In Chapters 5, 6 and 19 of my Bitemporal Data: Theory and Practice, I extend Grice's rules to the processes of inserting, updating and deleting data in databases, and to the querying of those databases by people looking for information. I also introduce speech acts as essential to the third temporal dimension I define in that book.

 

I think of my work in that book, and also of the new work on discourse representation, as providing detail and formalization to what Winograd and Flores wrote thirty years ago.

 

To return to the point I began with: useful work in any non-trivial area of research very seldom comes from those who know little or nothing about what other serious researchers have already discovered and formulated. (Aristotle and Plato are exceptions that prove that rule.)

 

Becoming conversant in a field of research is hard work. I'm in the process of extending my pretty good knowledge of ontology to ontology engineering, i.e. to the combination of ontology and logic that distinguishes this work from classical ontology. But spending a few hours, or even days, or even weeks and months, getting up to speed in any non-trivial area of research, is at best barely enough to let you catch the jist of what serious researchers have written. 

 

We all like "bull session" kinds of conversations, the ones that flow best face-to-face, usually late at night and after a few beers. In those conversations, there's no need to distinguish informed from uninformed talk about "deep stuff" like Reality, Mind, Language, Thought, Concepts, Relativism, Skepticism, Reductionism and Verificationism. 

 

But like John, Pat, Leo and some of the others in this forum most of whom have worked hard enough to not only have PhDs in related fields of study, but who have continued to deepen their understanding of their fields of study through a lifetime of work, I feel that bull session, "I don't know anything about Kant but I do know about Reality, Concepts, Objectivity, etc" conversations are not appropriate in this forum. Maybe on LinkedIn; but not here.

 

This is not to disparage anyone's interest in anything that requires hard work to master. But when uninformed people go on and on about these topics, in a forum supposedly devoted to serious conversations, it seems to me to show a real lack of respect for the hard work that the more informed among us have put in. It's like a freshman in a Philosophy 101 class trying to take over the class with a discussion of his "philosophy of life".

 

Let everyone here take part in conversations. In peer-reviewed journals, that freedom doesn't exist, and the consequences of that restriction are sometimes unfortunate. In open forums, like Linked In, it's the opposite -- a free-for-all, an on-line bull session. That has value as social and professional networking, but also, frankly, as a place for poseurs to pretend to knowledge and expertise they don't have -- and, usually, that they aren't even aware they don't have.

 

I think that the ideal role for the Ontolog Forum is as a balance of these opposites. Something like a graduate seminar, or at least like the best of them I have attended. Graduate students in seminars are expected to engage in serious conversations. But if the professor has the expertise he should have to be teaching the seminar, each student's contributions to that conversation, to that symposium (which was originally a drinking party!), should be expressed as clearly and completely as possible, but with acknowledgement of the professor's expertise and acceptance of his role as the most informed participant in the conversation. Acknowledging the professor (or professors', sometimes) role leads to a certain kind of behavior on the part of the students. And that behavior is not the behavior of going and on, session after session, as though anything the professor(s) had said had never been said.

 

Intellectual humility is a virtue. It's not the virtue of trying to convince yourself that you don't know anything. But it is the virtue which recommends listening more than talking when more informed people than yourself are willing to participate. 

 

I do wish this forum were more like a graduate seminar than a bull session.

 

With apologies for writing what may be offensive, but which I think needed to be said,

 

Best wishes,

 

Tom

 

 

 

 

 

 

On Wednesday, July 8, 2015 5:03 PM, Rich Cooper <metasemantics@xxxxxxxxxxxxxxxxxxxxxx> wrote:

 

Dan Ariely is a Duke Prof who explains the evidence about how people make decisions under various conditions.  This is  his talk:

 

https://www.youtube.com/watch?v=bGL_CWHP78Y

 

So it would seem that knowing these details, one could write a requirements doc about how the conversational interface would help people deal with these tendencies. 

 

Sincerely,

Rich Cooper,

Rich Cooper,

 

Chief Technology Officer,

MetaSemantics Corporation

MetaSemantics AT EnglishLogicKernel DOT com

( 9 4 9 ) 5 2 5-5 7 1 2

http://www.EnglishLogicKernel.com

 

From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx [mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of Rich Cooper
Sent: Wednesday, July 08, 2015 1:19 PM
To: '[ontolog-forum] '
Subject: Re: [ontolog-forum] Ontology based conversational interfaces

 

OK, John,

 

That's a constructive suggestion you made:

 

JFS: One of the few instances in which he says "the world" could be replaced by the phrase "planet earth" without changing the point:

 

    "Much of our experience of the world comes from inside our brains."

 

I don't particularly like stipulating that "planet earth" captures the concept, but since you don't like the word "world", let's not put in any substitution at all for the main point: its "objective reality" we should put there:

 

    "Much of our experience of objective reality comes from inside our brains."

 

Perhaps that will fly.  Is it OK with you?  Does anyone else object to the verbalities?

 

It's a reference to the stored experiences of reality, in our memories, from deep within our earliest sensations, which were the closest we could ever get to objective reality, until now.  In the duration, we have become more and more biased in our particularly chosen directions. 

 

Sincerely,

Rich Cooper,

Rich Cooper,

 

Chief Technology Officer,

MetaSemantics Corporation

MetaSemantics AT EnglishLogicKernel DOT com

( 9 4 9 ) 5 2 5-5 7 1 2

http://www.EnglishLogicKernel.com

-----Original Message-----
From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx [mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of John F Sowa
Sent: Wednesday, July 08, 2015 12:36 PM
To: ontolog-forum@xxxxxxxxxxxxxxxx
Subject: Re: [ontolog-forum] Ontology based conversational interfaces

 

Bruce, Ed, and Rich,

 

Many years ago, I learned that if some word is a cause of many confused and confusing arguments, it's a good idea to *banish* that word from the discussion.

 

Bruce

> What is common, what is world?  That, too, is a matter of stipulation

> and agreement.

 

Yes, indeed.  For that matter, the *only* precise meaning for the word is "planet earth".  Everything else is an extended use or metaphor that varies from one context to another.

 

Recommendation:  In every occurrence of that word in this thread, replace the word 'world' with a word or phrase with a narrower meaning.  If you mean planet earth, say so.  If you mean world view (or German Weltanschauung) say so.

 

Ed

> U.S. Republicans and Democrats, like the political parties of other

> major “democracies”, must agree on a “common world”

> (universe of discourse) in order to communicate and legislate.

 

Politics is an example where a huge number of problems are created by the choice of words.  All the parties could agree much more quickly if they avoided words such as 'freedom', 'amnesty', 'religion', 'conservative', 'liberal', 'socialist', 'extremist'...

 

Successful politicians are not stupid.  They don't use those words when they're talking one-to-one with no cameras around.  But as soon as the cameras are turned on, the discussion turns to mush.

 

Rich

> Dan Ariely explains some of the reasons why we see different worlds:

> https://www.youtube.com/watch?v=_Y0w5EJC9o0

 

That's a good talk.  But Dan A. does *not* use the word 'world'.

I took some notes, and he uses narrow, precise terms:

 

    "How would you design an experiment?"

    "How do you classify experience?"

    "The brain is filtering information in a biased way."

 

One of the few instances in which he says "the world" could be replaced by the phrase "planet earth" without changing the point:

 

    "Much of our experience of the world comes from inside our brains."

 

Recommendation to Rich:  You have been creating a huge amount of confusion in this thread by using the word 'world' in a hopelessly vague way.  If you want to continue discussing the topics in this thread, please *stop* using that word.

 

If you need help, go back to Dan A's talk and take notes on which words he uses.

 

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

 

 

 


_________________________________________________________________
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)

<Prev in Thread] Current Thread [Next in Thread>