Re: [xtm-wg] The Topic Naming Constraint

["Bernard Vatant" <universimmedia@wanadoo.fr>]

The debate around the TNC is very interesting indeed. But I think it missed
somehow so far the real core of the question.

The spec is as it is. It's surely very consistent from a computer viewpoint
:)
But there are and will always be ambiguities with names and identities. Not
because of anything broken with the spec. If there is a place where
something is broken, inconsistent, ambiguous, whatever, it's in the natural
language, the one which affects names to subjects, or, to be more accurate,
the one which invents names and then *pretend* they represent subjects and
sort them out.
This process is very ambiguous and every other subject is a moving target.
How can anyone with some common sense imagine that, by some magic, this
ambiguity of subject will disappear because we represent it as a topic in a
topic map ? There is no way to do that. A subject is not the same animal
that a computer object with a defined id, and it'll never be, even if we
pretend that by the reification process. The ambiguity will not disappear,
and indeed it will even appear very well ! The best we can hope, and it
always has worked this way, is that the new structuring tool acts back in
our use of natural language and pulls us towards more accuracy in it. It's
what science, logic, maths have done for centuries, tracking such moving
subjects as time, space, energy, measure, life ... and lawyers too,
tracking more moving and ambiguous ones like democracy, freedom, justice,
rights ...
The ambiguities and weird mergings we'll find coming out of the TM
processes are indeed very good things. They will learn us to be less
arrogant about our language and its crazy pretention to grasp the
complexity of universe.

That important metaphysical stuff being said. On a technical viewpoint : we
merge Topic A, with PSI1, with Topic B, with PSI2, out of TNC, and we have
now Topic C with two PSI which look inconsistent. What has happened ?
- If the two topics were created distinct, with different PSI, by the same
author considering them as distinct subjects, he should have managed to
give them distinct names. "Please say what you mean". Find a way, like in
natural language.
- Maybe it's because the subject is controversial. Suppose we have two
different teams developing a TMQL and each pretending they have "the" TMQL
... well, you can't ask the computers to settle that conflict, only to show
it off : TMQL is a controversial subject, they are two of them, pick your
choice ...
- If the topics were created out of some automatic process, well, there
again, it shows the inconsistencies of the stuff which was processed, not
of the process. If you try some semantic analysis tool on this text I've
just written and represent it as a TM, certainly some inconsistencies will
show off, pushing me to come back to what I meant and disambiguate it a
little further.

Bernard



------------------------ Yahoo! Groups Sponsor ---------------------~-~>
eGroups is now Yahoo! Groups
Click here for more details
http://us.click.yahoo.com/kWP7PD/pYNCAA/4ihDAA/2n6YlB/TM
---------------------------------------------------------------------_->

To Post a message, send it to:   xtm-wg@eGroups.com

To Unsubscribe, send a blank message to: xtm-wg-unsubscribe@eGroups.com 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/