RE: How namespace names might be used

>> I have never, ever suggested that always retrieving the resource should
>> be the expected behavior. ...
>
>But Mr. Connolly has (as has, if I mistake not, Mr. Berners-Lee):

I've lost track of who said what, but in fact TBL has been suggesting that
the URI itself should be the Namespace Identity.... and that retrieving
anything via that URI is a completely seperate matter.

(Tim has expressed some ideas on why it might be useful to do so, directly
or indirectly, but that's a seperate layer of processing. As far as what's
been defined so far, a Namespace has identity and _only_ identity, though
that identity may carry additional connotations such as "ownership" as a
side-effect of the use of URIs.)

This conflation of identity and retrieval has been a serious impedement to
understanding the question. If there's any chance we can get away from it,
let's do so. PLEASE.


I'm still not sure that URIs themselves are a good choice for namespace
identity. I _am_ pretty sure that making a relative reference to that
identity is a Bad Idea and should be discouraged or (preferably) forbidden
until and unless someone comes up with a really compelling use case that
gets us all to agree on what this is supposed to mean.

______________________________________
Joe Kesselman  / IBM Research

Received on Monday, 12 June 2000 14:38:46 UTC