[
Lists Home |
Date Index |
Thread Index
]
jonathan@openhealth.org (Jonathan Borden) writes:
>You might find "ontologies" to be about this, but perhaps that is
>because you are looking at "ontologies" with a limited viewpoint.
>Granted there are project to build a standardized ontology of the
>universe e.g. Cyc and SUO, but those particular projects are *hardly*
>all that ontologies in general are about.
I said I find "making claims of certainty about the universe [to be] a
project I find unfortunate". That definitely puts the rhetoric around
Cyc in the discard bin, but it seems more like a recognition of the
limits of our understandings than a limited viewpoint in and of itself.
>Linda has got it right, ontologies can be used as tools in getting
>projects done. Criticizing "ontologies" in general is akin like
>criticizing "mathematics" as a project that you might find unfortunate.
Mathematicians and ontologists both frequently fetishize logic and think
they're tapping into something grander than the rest of us can see. To
the extent that such projects can avoid that and remain focused on the
practical, I'm thrilled. I don't share their faith in logic or the
excitement I see for pinning knowledge on virtual index cards.
>Do you find the American College of Pathology's SNOMED project
>unfortunate? How about the U.K. Galen project? Do you find
>dictionaries or thesauri unfortunate? Do you have any respect for the
>library cataloging community? Dewey Decimal System? etc.
All of these things have their uses and abuses. Dictionaries and
thesauri, while wonderful reference, seem to provide reassurance to
self-styled dictators of the meanings of words. Library cataloging is
great as an aid to helping people find books (my mother did her MLS
thesis on comparative cataloging systems), but taking that as a step
toward a grand theory of knowledge makes me distinctly uncomfortable.
From what little I know of SNOMED, it sounds fine in the hands of
doctors, but I have worries about how similar information has been used
in the hands of insurance companies. I just don't know.
>You have the perfect right to wear as dark a shade of sunglasses as
>you like while walking around your own house, just as I might walk
>around my own house with both hands firmly tied behind my back. I'd
>find that rather limiting at work, and analogously some folks might
>find ontologies rather useful at work as well. The "chain of
>certaintly" or "claims about the universe" are just phrases that may
>or may not help us get our work done.
Unfortunately they're phrases that come up on a regular basis at
conferences. I'd actually like to talk about markup, not the value of
interlocking assertions made by blessed experts. You can call that
wearing sunglasses or blinders if you like, but to me it's focusing on a
set of technologies that I find useful.
If you want to spend your time developing ontologies in the privacy of
your own house, you're quite welcome to do so. If people come on
xml-dev and make claims that, for instance, ontologies are the right
answer to vocabulary development for XML, they should not be surprised
to encounter resistance from people who have other approaches.
--
Simon St.Laurent
Ring around the content, a pocket full of brackets
Errors, errors, all fall down!
http://simonstl.com -- http://monasticxml.org
|
