Peter P. Jones wrote:
> On 24 Feb 2004 at 1:32, Murray Altheim wrote:
> [...]
>>As you may have gathered from my input here, I am not one to deride
>>philosophical investigation, though my interest is more narrowly
>>focused on epistemological issues. Your three points would seem to
>>imply a causal, temporal relationship, but I'd say that part of the
>>reason to frame collaboration issues thusly is to enable them to be
>>looked at within a wider scope of inquiry, not because that framing
>>will provide (necessarily) the answers. I don't look at philosophy or
>>epistemology as providing answers, nor do I think we'd have to solve
>>any outstanding philosophical or epistemological issues in order to
>>move forward.
[...]
> So to get back to the collaboration system design theme, poor tools
> will make for poor craftsmanship whatever. However, it is clear in
> the above that it is very difficult to say that one can select an
> optimal epistemological stance for some project without impoverishing
> some aspect or another - especially given time limitations. (01)
The one point that I would reiterate is that while it may be difficult
or even impossible to select an *optimal* epistemological stance, it
seems that the contrary and typical approach, i.e., to ignore the issue
entirely, leaves the project open to differing interpretations. IOW, if
there is not *some* documentation about what the project designers mean
by various terms, design structures, etc., then this leaves this open
to varying, and differing interpretations. (02)
For example, unless I'm wrong, OWL jumps right in with "class", building
a substantial ontological system without ever defining what the term
means. It's defined in terms of OWL itself, but not in terms of what it
means in human terms and how it can be correctly applied to real-world
classes (whatever they mean by classes). Are these classes intensively
or extensively defined, what are their boundaries, are they platonic,
are they in any way contextualized (either internally or externally)? (03)
There are a few people, like Nicola Guarino and Aldo Gangemi, who pay
attention to these issues. Gangemi's background is not surprisingly in
philosophy. (04)
> The question is whether awareness of philosophical theories
> constitutes an effective part of the toolset for collaborative system
> design. (05)
Certainly. I would argue that one ignores it at one's peril and ignorance. (06)
> I would contest that an awareness of philosophical theories at least
> contributes an increase in the possible subtlety with which a project
> can be approached and adds to the toolset in that sense. E.g. it
> might oppose pernicious reductiveness in the design of some crucial
> OO modelling of an interaction to the extent that it informs the
> system designer's initial inquiry. (07)
I've considered that Ceryle is a relatively "safe" approach to
ontology because it is designed as a tool for an individual to
develop and use as according to their own ontological/epistemo-
logical viewpoint. It's not necessarily for sharing, though of
course nothing stops an author from doing so. But at the point
at which an ontology is shared among multiple parties, it's
important that the base assumptions, the fundamental vocabulary
and grammar of expression, be exposed, otherwise any ambiguity
will lead to varying interpretation, semantic mismatch/error.
This cannot be eliminated in human communication, only minimized
via documentation, but hopefully the designer/author has thought
through these issues enough to provide documentation. If not,
then *everybody* is in a muddle (e.g., if you don't know what
constitutes a class, how can you reason about it?). (08)
Murray (09)
......................................................................
Murray Altheim http://kmi.open.ac.uk/people/murray/
Knowledge Media Institute
The Open University, Milton Keynes, Bucks, MK7 6AA, UK . (010)
Now the Pentagon tells Bush: climate change will destroy us
http://www.guardian.co.uk/climatechange/story/0,12374,1153530,00.html (011)
--
This message is archived at: (012)
http://collab.blueoxen.net/forums/cgi-bin/mesg.cgi?a=yak&i=403B55DF.30501@open.ac.uk (013)
|