[topicmapmail] Re: paradigmatic PSIs

[Lars Marius Garshol]

* Ann M. Wrightson
| (small extension of mail previously sent to sc34wg3)
| 
| "The paradigm" (as far as we can judge from its apparent phenomena) appears
| to assume that:
| 
| Subjects are atomic
| Subjects are identifiable
| Subjects are persistent
| Subjects have no dependency on domain, perspective, user, etc etc

I love this! That's an excellent summary of the assumptions topic maps
have made about subjects, without really spelling them out. Wonderful!
 
| IMO the trouble here is exactly that "The paradigm can't really
| handle *all* subjects, without any exceptions". It's (just) possible
| that some interestingly large community will succeed, using the
| paradigm. It's quite likely that many communities of useful size
| will succeed - by virtue of their existing (implied or express)
| consesnsus regarding subjecthood, or by using an agreed set or
| scheme of subjects. Hope for wider kinds of universality is
| admirable but hopeless.
| 
| Why?
| 
| The identification of a subject is *relative to the intended use of
| the identified subject* as has been found by many developers of
| thesauri and classification schemes. Further, adequate
| identification of a subject can be a distinctly non-trivial matter
| (even though lots of them, such as "opera", "aria", "nuclear
| assembly", come ready-identified within a community sharing enough
| common language and culture).

I agree completely with what you say here, but I don't think that
means a failure for topic maps. Some people in the topic map community
want to do "global knowledge federation" with topic maps, but many
people (such as myself) don't care about that at all, or even that it
is possible. 

Personally, I'm only interested in using topic maps in more limited
contexts where this is much less likely to be a problem. I think there
will certainly be cross-usage of published subjects, and that there
will be a "democratic" or "evolutionary" process where certain PSs
emerge as "winners", but that there will not be universal agreement on
what subjects to use.

So if what you say is true it means that "global knowledge federation"
is not achievable, but it doesn't make topic maps unusable or anything
like it.
 
-- 
Lars Marius Garshol, Ontopian         <URL: http://www.ontopia.net >
ISO SC34/WG3, OASIS GeoLang TC        <URL: http://www.garshol.priv.no >