At 09:47 AM 3/19/2008, John F. Sowa wrote:
>I certainly agree with that point:
> > If an OOR is to be useful, then potential users need to have
> > reliable expectations as to what it will contain.
>But the next assumption does not follow:
> > This means some sort of gatekeeping. This means in turn that
> > some things will not be included. But not including something
> > in an OOR is not equivalent to banning that something. There
> > are always other ponds in which one can play.
>An important point about a repository is that it contains a
>large amount of metadata, among which would be records about
>who developed, used, revised, and extended it and the results
>that were obtained in various applications. (01)
Any old metadata? In any old format? In Gaelic? (02)
>But an ontology that is optimal for one range of applications
>is not likely to be optimal for all purposes. Therefore, it's
>essential to include all the alternatives, with a detailed
>record of their uses, successes, and failures. (03)
Any old record? In any old format? In Gaelic? (04)
>provides the guidance that developers need to make a choice. (05)
Should it be ... honest?
If people are caught supplying dishonest metadata should they be banished? (06)
>Then they can "vote with their feet" to determine which one(s)
>are most appropriate for various applications. And the ultimate
>decisions would be based on results -- not a priori opinions. (07)
All my opinions are a posteriori.
I assume that yours are too. (08)
>At this stage, there is very little evidence that anybody knows
>what an ideal ontology would be. (09)
I seek not an ideal ontology, but a good ontology, which I can
recommend in good faith to ontology non-aficionados.
Msg Archives: http://ontolog.cim3.net/forum/ontology-summit/
Community Files: http://ontolog.cim3.net/file/work/OntologySummit2008/
Community Wiki: http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2008
Community Portal: http://ontolog.cim3.net/ (011)