Message-Id: <9402281753.AA24845@expresso.bunyip.com>
From: Peter Deutsch <peterd@bunyip.com>
Date: Mon, 28 Feb 1994 12:53:53 -0500
In-Reply-To: Martin Hamilton's message as of Feb 28, 12:21
To: Martin Hamilton <M.T.Hamilton@lut.ac.uk>, mitra@pandora.sf.ca.us (Mitra)
Subject: Re: URN to URC scenario
Hi Martin,
[ You wrote: ]
. . .
> It occurs to me that we might be making this whole naming authority/
> publisher business a bit too complicated - is there any support for
> merging the naming authority and the publisher info? i.e. instead
> of URNs being composed of at least
>
> naming authority + publisher info + resource ID
>
> this would be transmuted into
>
> publisher info + resource ID
Actually, I'd prefer to see it become:
naming authority + resource ID
With the publisher ID subsumed perhaps into the opaque
string of the resource ID.
The way I see it, the publisher info may or may not be
appropriate (eg. some schemes wont have a specific
publisher ID), but the naming authority is needed to allow
us to experiment with more than one scheme and we'll
definitely want to be able to do that.
Some schemes will contain DNS hooks for server location,
some will be easily computable, some will be more easily
readible, some will be strings of numbers, etc. The naming
authority allows a client author to write code to decide
right away if this is something it can handle, or needs to
bounce/hand to a proxy server/whatever. The actual
dereference is the second step that follows this first one
in this world view. Without the naming authority I don't
see how you can decide what to do next.
- peterd
--
---------------------------------------------------------------------------
"The future belongs to neither the conduit or content players, but
those who control the filtering, searching and sense-making tools
we will rely on to navigate through the expanses of cyberspace."
- Paul Saffo, (_Wired_: March,1994)
---------------------------------------------------------------------------