Re: Web Resource Identity

From: Henrik Frystyk Nielsen (frystyk@w3.org)
Date: Sun, May 30 1999


Message-Id: <3.0.5.32.19990530141857.0302e970@localhost>
Date: Sun, 30 May 1999 14:18:57 -0400
To: David Megginson <david@megginson.com>, Paul Prescod <paul@prescod.net>
From: Henrik Frystyk Nielsen <frystyk@w3.org>
Cc: xml-dev <xml-dev@ic.ac.uk>, lavoie@oclc.org, www-wca@w3.org
Subject: Re: Web Resource Identity

At 14:23 28/05/1999 -0400, David Megginson wrote:
>  "Are these known to refer to the same object?" => Yes/No
>
>And while a cononical URL is an easy extension to HTTP, the
>infrastructure to support it is another issue -- do I have to
>guarantee, for example, that a gzipped tar file of the same version of
>the Linux kernel source at two different mirrors return the same
>canonical URL?

Note the distinction between a resource manifestation and the resource
itself. Proving that two URIs point to the same resource where "same" can
be arbitrarily abstract is one thing (and what I think Paul's first
question was about). Proving that you got the same bag of bytes is another
- a secure hash would do the trick.

Henrik
--
Henrik Frystyk Nielsen,
World Wide Web Consortium
http://www.w3.org/People/Frystyk