W3C home > Mailing lists > Public > public-prov-wg@w3.org > June 2011

Re: Web architecture and metadata

From: Simon Miles <simon.miles@kcl.ac.uk>
Date: Thu, 2 Jun 2011 15:41:00 +0100
Message-ID: <BANLkTim5a9hsVfofeSQHagBBEEo1XnPDig@mail.gmail.com>
To: Provenance Working Group WG <public-prov-wg@w3.org>
Graham, all,

I realise that there have been many posts since the one I'm replying
to, but the articles you suggested (esp. [1], [2]) seems directly
related to, and agreeing with, what Jim and others have been saying,
and I haven't seen anyone related their arguments to our discussion.

In brief, I think that their arguments would support proposals for us to:
 - talk only about resources (not states, representations etc.) for
anything of which we could find the provenance, and
 - first define what the provenance of a resource is with regards to
just those aspects for which it is considered immutable, before
loosening this assumption for user convenience

In less brief:
An implication of their arguments are that distinctions between web
"resource", "state" and "representation" are not precisely defined and
anyway merely examples of a wider range of
generalisation/specialisation relationships. This seems to support the
suggestions (from Graham and Martin?) that we do not distinguish
resource and resource state: everything's a resource.

I would express the main argument of the articles in my own terms as:
we can talk about the metadata of any resource *in those aspects for
which it is immutable* (or, in their terms, where it is true for every
specialisation of the resource). So the Royal Society, while
definitely a mutable resource, will always have been founded in 1660
[3]. A statement of its membership however, would have to apply to the
Society as it is on a particular date, a specialisation of the first

[1] additionally comments on mapping this abstract idea to practice on
the web. Specifically, a general resource may be referred to when you
are actually making a statement about one specialisation of that
resource if it is clear from context which specialisation you mean,
e.g. your statement uses the URI of a webpage but actually refers to
the serialisation of that webpage in your browser.


[1] http://www.w3.org/2001/tag/awwsw/ir/20110517/
[2] http://www.w3.org/DesignIssues/Generic.html
[3] http://en.wikipedia.org/wiki/Royal_Society

On 31 May 2011 15:20, Graham Klyne <GK@ninebynine.org> wrote:
> I just spotted this work-in-progress of the W3C TAG, which might have some
> bearing on our approach to provenance...
> http://www.w3.org/2001/tag/awwsw/ir/20110517/
> This in turn leads (directly and/or indirectly) to:
> http://www.w3.org/2001/tag/group/track/issues/57
> http://www.w3.org/2001/tag/group/track/issues/62
> http://www.w3.org/2001/tag/group/track/issues/63
> http://www.w3.org/2001/tag/2009/02/metadata-survey.html
> http://odontomachus.wordpress.com/2011/02/21/the-place-of-metadata/
> http://www.w3.org/DesignIssues/Generic.html
> It appears this is a live issue for the TAG:
> http://lists.w3.org/Archives/Public/www-tag/2011May/0086.html
> #g
> --
> ______________________________________________________________________
> This email has been scanned by the MessageLabs Email Security System.
> For more information please visit http://www.messagelabs.com/email
> ______________________________________________________________________

Dr Simon Miles
Lecturer, Department of Informatics
Kings College London, WC2R 2LS, UK
+44 (0)20 7848 1166
Received on Thursday, 2 June 2011 14:41:28 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:58:05 UTC