Re: [geolang-comment] ISSUE 3: The "language" published subject

[Steve Pepper <pepper@ontopia.net>]

At 12:07 02/05/02 +0200, Lars Marius Garshol wrote:

>It is given that we will create one published subject for each code in
>the ISO 639 standard. The question is: should we create a "language"
>published subject?
>
>This is an issue because it seems likely that most people will want to
>assert that these published subjects are instances of a class
>"language", but unless there is a published subject for this concept
>these assertions will not merge correctly.
>
>(Of course, some people may not want to, but the existence of a
>"language" published subject will not force them to use it, even if
>they use the individual code subjects.)
>
>So, should we create such a published subject?

Yes.

>If we do, how do we define "language"?

If ISO 639 provides an intensional definition, we should use that. 
Otherwise we should give it an extensional definition based on ISO 639: 
"The class containing all subjects for which there exists a code in ISO 
639", or some such.

>If we don't, do we just accept that every user of language.xtm will
>most likely define their own "language" topic?

Like you said: Nobody *has* to use the language PSI we provide if they 
don't want to... unless of course we publish the PSI set in a formal 
notation (such as XTM or RDF) and make the assertion for each subject that 
it is an instance of the class "language", and I don't think we should do that.

Steve

--
Steve Pepper, Chief Executive Officer <pepper@ontopia.net>
Convenor, ISO/IEC JTC1/SC34/WG3  Editor, XTM (XML Topic Maps)
Ontopia AS, Waldemar Thranes gt. 98, N-0175 Oslo, Norway.
http://www.ontopia.net/ phone: +47-23233080 GSM: +47-90827246