Re: [geolang-comment] First proposals for ISO 639 and 3166 available

[Bernard Vatant <bernard.vatant@mondeca.com>]

*Murray

> "710" is not an information resource relevant to the country South
> Africa, it is a coded name for that country, the context (or scope)
> of that name being within the set of codes as published by the UN.
>
> I think the XTM specification is starting to be read like the bible,
> wherein one can seemingly prove anything. In this instance it should
> be remembered that occurrences in the topic map paradigm are meant
> to be out in the world, not in the map. Names of topics reside in
> the map, and "710" is an alternate name for "South Africa", not an
> occurrence of a resource about South Africa. If we begin populating
> our topic maps' occurrences with things that should be in the map,
> where will the "real" occurrences live? The clean separation between
> map and territory mapped is lost.

As Lars Marius, I tend to think that what you call clean separation is actually a wide
fuzzy non-boarder zone, of the same kind indeed that the ones you point out rightly in
your other message to exist between languages. BTW if genetics have happily ruled out the
notion of race (even if too many people are not quite aware of that), the notion of ethnic
groups still holds I think, but like attractors which are mix-up of cultural, linguistic
and genetic common pool. And those have fuzzy non-boarders too.

What is the boarder between long descriptive names like "Republic of South Africa" and
short descriptions like "The main country in Southern Africa"? The fact that one is
trapped inside the topic map as a name, and the other stored in an external file as an
occurrence? That does not make much conceptual difference to me. Relaxed the Topic Naming
Constraint, neither will be used as identifier, and both can be considered as labels, or
as occurrences, depending only on where they are physically stored ... which users don't
care basically.

Other detail : "710" is not an occurrence by itself, it's just a resource data that
happens to be linked by a given type of occurrence (UN code) to the topic "South Africa".
If that code what stored in an external data base and retrieved by a resource reference
like http://www.uncodes.org/search?q=South+Africa would it change your view on that?

Bernard