[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
CE RE: Natural Language RE: Re: Now What?
Bill,
. As discussed, I am posting this item.
. Perhaps you are aware there is a "CE" site at suo-ce@ieee.org
<mailto:suo-ce@ieee.org> , and that's where I have posted this. That site
is purely for people interested in this topic.
-----Original Message-----
From: Horn, Graham
Sent: Monday, June 04, 2001 11:28 AM
To: 'William Burkett'
Subject: RE: Natural Language RE: Re: Now What?
Bill,
. Further comments interspersed below, prefaced "GH>".
. With your permission, I'd like to post this.
Cheers Graham Horn
National Data Standards Unit
Australian Institute of Health and Welfare
===========================================
Phone: +61.2.6244.1094
Fax: +61.2.6244.1199
E-mail: Graham.Horn@aihw.gov.au
-----Original Message-----
From: William Burkett [mailto:WBurkett@PDIT.com]
Sent: Saturday, June 02, 2001 1:23 AM
To: 'Horn, Graham'
Subject: RE: Natural Language RE: Re: Now What?
Hi, Graham,
> . I don't understand what you mean by "semantically organic".
To answer this question, I think I'd best state some assumptions/opinions
that underlying my comments so that you understand my biases. After many
years of working with and thinking about data models and data, I am growing
firmer and firmer in my belief that people use data like they use speech -
Data is Speech. It's subject to the same ambiguities and shortcomings as
natural language; it is as sensitive to context, intention, and nuance as
speech is. More importantly, it *evolves* like natural language
organically. Language evolution is a natural process and data/schemas are
caught up in it.
GH> Agreed. This is one of my reasons for believing we should be using a
modified & restricted English, which is far more intuitive to ordinary
people.
> . Do you mean that the structure should be
> logically sound, so
> that it copes with all logical possibilities?
In a sense, yes. A sound structure is important for adaptability and
evolution. But - and this will be a bone of contention with the SUO - I
don't think "logical" soundness is a practical requirement. People and
language are not logical, much as we'd like to think they are and want them
to be.
GH> I guess my belief is that language has evolved over millennia by
human beings assembling structures on a trial & error basis. Some of these
structures have turned out to sound pieces of logic, and others are
approximations, a bit like the musical "well tempered Clavier", ie practical
to use in many situations. As with many technologies, we have come a very
long way. However, I don't believe we have found any fundamental unifying
logic, much as we haven't found the fundamental subatomic particle, or
definitively identified the nature of the circumstances around the supposed
"big bang" of astronomy.
I don't want to pretend like I have a solution to all this in my head - I
don't. But I have seen FAR far too many grandiose schema development
projects with lofty, world-integrating objectives end up as efforts in
futility, wastes of money, and thousands of pounds of paper shelfware.
GH> Agreed. This is why I am proposing a relatively modest structure,
building on a structure developed over the last several millennia. It also
means we can benefit from the development work already done over that
period.
Why don't any of these work, I keep asking myself? Why can't this problem
be solved? The answer, I think, is that we're looking in the wrong place
for the solution.
GH> Perhaps I'm a bit more pessimistic. I'm not sure there is such a
lofty structure. Another reason for using something to hand which
demonstrably works, and extracting a logically sound subset (and possibly
making a few logical modifications and/or extensions). We can add the pieces
of logic identified by philosophers and logicians in terms of this logical
structure, and thereby get our SUO.
Bill