Amanda, (01)
First, let me say that your email is a "keeper". I find myself agreeing
with almost all of it. In brief:
- We need to spend our time building real ontologies that solve real
problems
- We can use upper ontology concept sets where they help in supporting
and elucidating the models we make. (We tend to re-create these concept
sets on our own when we need them; better is to reuse them.)
- We need to have good means of access to the concept sets in proposed
upper ontologies, so that we can reuse them.
- In this way we can learn what upper ontologies and concept sets are
most useful, and perhaps to what purposes, and upper ontologies that are
useful will "emerge". (02)
I hope I got these tenets right. I would naively identify these as the
properties of a social system. (03)
In the course of your email, you mentioned two other things that I think
are important, but more in the background: (04)
> Within a small-scale, well-defined domain, it's often the case that little
> or no upper ontology is needed. The need comes when we want to cross
> contexts, however defined. (05)
Yes, but even small domains reuse certain patterns: part and grouping
relations, temporal relations, and the list you give below: Event and
Process, Individual and Organization, Conceptual and Physical, and
Information. These are areas in which you need reference models to
support automated reasoning about the domain, and the domain experts
(SMEs) take some such model as given. So you don't want to burden the
SME with the model (unless you can't figure out which one s/he is
using), but you still need one. (06)
> When a well-defined, context-specific ontology project occurs with
> cross-domain interoperability in mind, part of the task becomes the
> inclusion and/or connection with of defining concepts and relations that
> possible non-domain users would recognize. Middle-to-upper ontology
> questions come in quickly here, and naturally. (07)
Yes. (08)
> SMEs understand and can
> supply clarifying relations to such cross-domain concepts as Event,
> Process, Place, Time, Individual, Organization, Information, Physical
> thing. (09)
No. They understand, but they are often unable to verbalize the
concepts they have. And there are critical variations. I believe this
is only true in the light of what follows: (010)
> Supporting ontologists can ask the questions to identify what
> well-defined concept a SME is invoking. Ontologists are better able to ask
> the right questions and specifiy the right concepts because an experienced
> applied ontologist is a sme about these mid-to-upper concepts. The
> ontologist is therefore able to *ask the right questions*: the questions
> that distinguish the upper categories from each other, define the upper
> relations. (011)
Yes!, but there is more to this. The ability to deal with SMEs and "ask
the right questions" is a skill, and that skill is different from the
ability to construct a cohesive ontology from the answers. I don't know
how to teach this skill. I have worked with other knowledge engineers,
some of whom have it and some of whom don't, and there doesn't seem to
be a correlation between the ability to extract the information and the
ability to use it well in designing ontologies. And those who are weak
in this area invariably insist on imposing some predefined scheme on the
conversation with the domain expert, because they can't understand
his/her model as s/he expresses it. So some of the "imposition from on
high" that one sees is, and may be intended as, a crutch for the
knowledge engineer who is handicapped in this way. (012)
(This may be one of those things you had to learn as a baby, dealing
more with language than with things, so you learn by listening more than
by doing, or vice versa on both accounts.) (013)
Amanda later takes up this theme, but with a different instrument: (014)
> It does
> mean a certain tolerance of complexity, of pockets of unknownness. It does
> mean that we may ourselves not be able to see how all the different
> perspectives fit together; that's a kind of unknownness that some people
> find intolerable or equate with relativism (in the sense that is opposed to
> realism). (015)
It is important to realize that a difference in viewpoint can produce an
importantly different "factorization of the problem space". If you look
at an image of the earth in non-visible light spectra, you may not
recognize any element, even though the thing itself is the same. Or,
you may recognize some elements roughly, but not in fine, and not in
relationship to others. This kind of thing regularly occurs in the
perspectives of different SMEs on the same business activity. Amanda
hints that this is not "relativism", and she is quite right. This is
"reality" in someone's understanding, but that understanding is
"correct" and may be intrinsic in the thing/situation and is shared by
other SMEs who have the same background or the same concerns. (016)
And in many of these situations, we cannot build a single model of the
elephant that meets the needs of all the viewpoints. It is not that the
ear is the fan and the body is the wall; it is that the body is a wall
and is soft and permeable and generates heat and moves, and so does the
ear, but it is not a wall. ;-) (017)
And last, but not least, the Amanda Vizedom aphorism:
> Ontology is fractal. (018)
I have to think about that. :-) (019)
-Ed (020)
--
Edward J. Barkmeyer Email: edbark@xxxxxxxx
National Institute of Standards & Technology
Manufacturing Systems Integration Division
100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528
Gaithersburg, MD 20899-8263 FAX: +1 301-975-4694 (021)
"The opinions expressed above do not reflect consensus of NIST,
and have not been reviewed by any Government authority." (022)
_________________________________________________________________
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
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (023)
|