David, Mike, et al,
I think the lees are settling now. Many of us will agree upon having
one statement for a concept. And, to have one statement of a concept
entails that the statement must be understood by all in the target
community.
For me, that means there must be a canonical language for every
concept statement. And that is why many linguists have been
interested in a vocabulary for canonical speech (I hesitate to call
it an upper ontology because of the baggage).
To develop a canonical vocabulary, it is insufficient to follow
someone around all day and write down their words. If you do this,
you would miss words associated with the events of other days.
Instead, you must follow a number of people around for a year and
listen to their "sentences". Linguists begin with sentences in the
process of understanding a language.
Next, you must select the set of sentences that cover all important
situations. For this we need to embrace the concept of 100%
linguistic coverage. It is certainly possible to achieve 100%
coverage in syntax if we identify and assign a part of speech
identify to all words. For semantics this is more difficult but I
believe it is still possible. Austin's Speech Acts identify those
semantic functions that accomplish a physical "something". There
should only be a few other "Acts" that he missed to be identified.
Selecting the semantic functions that are important means that some
are more important than others. This means there must be an
estimation process to identify those situations that are important.
And those are defined by the environmental niche of the inhabitants.
Speech is only defined within its environment. However, we must
recognize that some of those "important situations" involve contacts
with foreigners. How do we communicate with them and use that
communication to establish a common language? It seems to me that
the members of the community are the ones who can tell us what the
important semantic functions are.
Some argue that this is the point where meta-linguistics becomes
important. Others argue for ostension and methods. Whichever, or
whatever other method, you prefer; we have not had that discussion
here. I don't see another way forward and the same has been true for
others. That is why, god help us, there is so much interest in
statistical gossip machines. There must be a better way.
And that is why I asked earlier how English (or another language)
could be partitioned. The general consensus seemed to be that it is
not possible. I disagree and you might want to ask that question of
a 4 year old.
Voice of America addressed this issue in developing their content
material and it works very well with BE1500. While this does an
admirable job, I believe it can be improved upon for computing
systems with a little intelligent thought.
-John Bottoms
First Star Systems
Concord, MA
On 2/22/2012 12:52 PM, Mike Bennett wrote:
On 22/02/2012 15:52, David Eddy wrote:
Mike -
On Feb 22, 2012, at 8:54 AM, Mike Bennett wrote:
it
became clear that what they needed, after various false
starts involving XML standards, common data models and
so on, was some resource in which each meaning was
defined once.
I'm having a hard time with "each meaning was defined once."
Aren't a meaning & a definition the same thing?
I should have put "Each meaningful concept is defined once". Does
that make it clearer?
--
Mike Bennett
Director
Hypercube Ltd.
89 Worship Street
London EC2A 2BF
Tel: +44 (0) 20 7917 9522
Mob: +44 (0) 7721 420 730
www.hypercube.co.uk
Registered in England and Wales No. 2461068
_________________________________________________________________
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)
|