RE: What Name Should A Namespace Name?

<- Anything URI that has a non-dereferencing scheme, but has a central
<- administration body.

Hmm...not sure I like the sound of that kind of org. There's also all the
copyright type chaos to deal with, which has to some extent settled down
with URLs.

That'll be a URN then. Having said that, I don't
<- see why a data:, URL couldn't be used, due to the level of
<- specificity. IMHO, a tann: [1] would be ideal for this... naming an
<- abstract concept directly without any particular resolution mechanism,
<- although people are by now so dependent upon HTTP, it makes me wonder.

I hadn't seen tanns before - neat idea (doesn't need the central admin for a
start), but perhaps too late.

<- That's why I was rambling a bit about using HTTP for namespaces - yes,
<- all of this has been covered in depth before, and there have been
<- endless circular debates, but these often end up disappearing down the
<- theoretical rat-hole. The fact of the matter is, namespaces are being
<- created all the time, as are terms, and we need to have some semblance
<- of order - best practises and the like, to ensure we don't end up with
<- a semantic melee.

I'm not sure how much order can be 'imposed' - a lot of areas are pretty
much tied to existing specs or common (rarely best) practices. I suppose
it's a matter of getting the taxonomy of low-hanging fruit right really

Received on Wednesday, 11 April 2001 14:16:39 UTC