Date: Wed, 14 Jul 93 15:16:44 EDT
From: rodgers%billings@billings.nlm.nih.gov (Richard Rodgers)
Message-Id: <9307141916.AA01609@billings.nlm.nih.gov>
To: krol@ux1.cso.uiuc.edu, masinter@parc.xerox.com
Subject: Re: Registrar
> From uri-request@mocha.bunyip.com Wed Jul 14 15:06:21 1993
> To: krol@ux1.cso.uiuc.edu
> Cc: raisch@ora.com, timbl@nxoc01.cern.ch, ccoprmm@oit.gatech.edu,
> terry@ora.com, uri@bunyip.com, www-talk@nxoc01.cern.ch
> Subject: Re: Registrar
> From: Larry Masinter <masinter@parc.xerox.com>
> Sender: Larry Masinter <masinter@parc.xerox.com>
> Fake-Sender: masinter@parc.xerox.com
> Date: Wed, 14 Jul 1993 10:28:06 PDT
> Content-Length: 415
>
> It may be that you can't get away with having one kind of URN. Why not
> have both 'mutable' URNs and 'immutable' ones?
>
> The mutuable ones refer to 'the latest version of this thing, as
> determined by the author', while the 'immutable one' refers to 'this
> is as good as if I had put away a copy'.
>
> When you assign a URN, assign two. When you update it, the update
> keeps the mutable URN and gets a new immutable URN.
>
Isn't this just equivalent to having version numbers of an entity?
I thought that we had decided that this was the sort of information which
would go into some descriptive material to be somehow attached to the
entity, but that it was not in general a good idea to elevate this to
the level of the URN. Perhaps, though a boolean flag somehow associated
with the URN is a good idea, one which simply states whether the document
is mutable or not; in the former case, one would be warned to look for
a version identifier if this is important for the application at hand.
In any event, it seems a poor idea to specify that URNs refer to mutable
documents, as no one can ensure that this will be the case...
Sorry not to be with you in Amsterdam--hope the deliberations are
proceeding with the usual alacrity! ;^)
Cheerio, Rick