Re: Versioning

[John Cowan <cowan@mercury.ccil.org>]

James Clark scripsit:

> >         urn:publicid:-:OASIS:NOTATION:TREX:1.0

I meant to write:  urn:publicid:-:OASIS:NOTATION:TREX:EN:1.0
where EN=English, the language of the TREX specification.

[snip]

> I think using a URN is good, and I think having the last component of
> the URN be the version number is good, but I don't see the benefit of
> using an FPI-based URI: it looks over-complex to me.

Except for the "NOTATION", which could just as well be "NAMESPACE" (the
currently proposed publicid URN syntax extends the dozen-plus FPI
document classes to be any arbitrary identifier), I think this
says everything that needs to be said: the document owner is OASIS
(a non-registered name), the name of the thing being labeled is TREX.

> What URN schemes are there available?

There are currently *no* general-purpose URN namespaces available.
Currently we have the "ietf" namespace, for RFCs and their close relatives;
the "pin" namespace, for Whois objects; and the "issn" namespace,
for serials (newspapers, magazines, etc.).  None of these is suitable.

-- 
John Cowan                                   cowan@ccil.org
One art/there is/no less/no more/All things/to do/with sparks/galore
	--Douglas Hofstadter