[topicmapmail] Latest RM?

[Murray Altheim]

Peter P. Jones wrote:
> On 30 Oct 2003 at 3:41, Murray Altheim wrote:
> 
> [...]
> 
>>Since I made public my reservations about the RM, I should also say
>>that I share Jack's enthusiasm for where it is heading. When I said
>>that I thought that it was "heading into KR territory", that is
>>because quite a number of the relationship types in my opinion look to
>>be, and *should* be, based upon a logical core. And I should also
>>state that it would please me to no end if the RM *could* support KR
>>(I've been working towards that goal myself for several years now, as
>>some of you know).
> 
> Hi Murray,
> 
> Could you elaborate for me (and maybe others) what you mean by 
> logical core as the basis for relationships?

I don't know if it's my week or what, but on separate lists this
week, almost every question has been one of "could you please
elaborate on ___", where to do so would take many hours of time.
I should obviously write a paper on this one.

There are already logical models for relationships, many such models.
There's one in predicate logic, there's one in Conceptual Graphs
(where its 'relations' are based in first order predicate logic),
there's one in KIF, one in DL, one in Cyc, one in frame systems, etc.

I'll reiterate what I wrote in a private email earlier today, though,
which is more of a caution than anything else in beginning to believe
the there *is* such a thing as "Knowledge Representation", that
expressions of human knowledge can be somehow be accurately and
unambiguously formalized according to some "logic". This would assume
that human knowledge, human communication/expression is itself formal,
itself logical, and *capable* of being formalized. You have to make a
big jump of faith to believe that, and empirical observation tends to
belie the notion.

What I believe Brandom has been saying is that KR doesn't exist. I
agree with him. What exists is "Knowledge Expression". I think that's
what the RM is trying to do. In its assertions we see statements
-- expressions -- of human "knowledge" (avoiding for now travelling
down that road).

So rather than try to create a formalism for the underlying knowledge
itself (which I think is a chimera), what we need to concentrate on
is a formalism for making statements. Interestingly, that is precisely
what Peirce's "Thirdness" tries to do, what his Existential Graphs
began to do. The brilliance of EG was that it avoided having to tackle
the impossible, settling instead on the attainable by formalizing the
mode of expression, not some core notion of was being expressed (if
that is even available for formalization).

 > Is the translation between assertion types and a logical
 > representation not straightforward?

I don't think there's any demonstrable relation of that type that
can be proven, except via faith. It's in that chasm between the
real world and mathematical models of it. We can avoid that by
creating a model for the expression model, i.e., a way of formally
making statements (assertions), absent the "logical representation".

Murray

......................................................................
Murray Altheim                    http://kmi.open.ac.uk/people/murray/
Knowledge Media Institute
The Open University, Milton Keynes, Bucks, MK7 6AA, UK               .

   Monkeys use thoughts to control robotic arm
     http://www.sfgate.com/cgi-bin/article.cgi?file=/c/a/2003/10/13/MN2018.DTL
   Bush uses media expertly to push apocalyptic view
     http://truthout.org/docs_03/091403J.shtml