Message-Id: <9310170407.AA15326@expresso.bunyip.com>
From: Peter Deutsch <peterd@bunyip.com>
Date: Sun, 17 Oct 1993 00:07:07 -0400
In-Reply-To: Larry Masinter's message as of Oct 16, 14:54
To: Larry Masinter <masinter@parc.xerox.com>, uri@bunyip.com
Subject: Re: The URN: wrapper and URLs...
[ You wrote: ]
> Mitra wrote:
> > Personally, I think I'd vote the (no URL: yes on UURN:), ...
>
> and Simon Spero replied:
> > I think this seems to be the general consensus; it doesn't affect the existing
> > base, allows URNs to be easily recognised, and can easily be generalised for
> > other resource identifiers.
>
> However, I don't think I agree that the URN: prefix is useful, and
> think it can be harmful if you want to add new kinds of naming
> authorities.
Whether we agree that the URN prefix is useful or not, the
current proposal does have a method for extending the set
of naming authorities easily. I think its somthing like
like "URN:naming_authority:publisher:id". The prefix is
to allow automated detection of the URN, the naming
authority is to allow extensibility.
> `Consensus' usually means that everyone agrees.
Actually, to be fair, the system requires only "rough
consensus". Thus, if there were only a few holdouts, it
would be perfectly acceptable to forge ahead. Having said
that, I don't think we yet have consensus, since I do see
a number of key players arguing for each point of view.
Perhaps we can agree to disagree at Houston...
> It seems short-sighted to assume that there will be only one style of
> URN; rather, subsuming the ISO object identifiers and ISBN numbers
> would be a good start.
I would see this as something like:
URN:iso-obj:...
and
URN:isbn:...
Oh, and somebody wanted brackets, so...
<URN:iso-obj:...>
and
<URN:isbn:...>
- peterd
--