W3C home > Mailing lists > Public > www-tag@w3.org > May 2011

ISSUE-57 & ISSUE-63 draft: Information Resources and Web Metadata

From: Jonathan Rees <jar@creativecommons.org>
Date: Tue, 17 May 2011 21:05:13 -0400
Message-ID: <BANLkTi=qLBsPV15dTawUa-d3AnGP8ccSyg@mail.gmail.com>
To: www-tag@w3.org
http://www.w3.org/2001/tag/awwsw/ir/20110517/ is the first of two
drafts relating to ISSUE-57 for us to discuss at the June F2F. The
second, which attacks ISSUE-57 directly by listing current and
potential solutions, should be ready next week. I hope all TAG members
(and others) will read them and comment.

Apparently I had no action to do this, although I think it's common
knowledge I've been working on it. Noah, if you'd like for me to
assign myself an action to do this, for agenda purposes, I'd be happy
to.

This draft also relates pretty directly to ISSUE-63 (Metadata
Architecture) and thus to ACTION-282 (draft a finding on same).
However it is much narrower than metadata architecture generally and
therefore does not satisfy ACTION-282.

Keep in mind that there is a third imagined document, which would be
the output of a community discussion framed by the first two
documents. That's a ways off.

-Jonathan

---------
Abstract

This note considers the semantics of metadata in which the subject of
the metadata (the "data") is specified using a URI that may be
dereferenced on the Web. This situation is complicated in that agents
might obtain different information on different dereference
operations, raising the question of what metadata is true or not of
the metadata subject.

It is proposed that the practical purpose of the "information
resource" abstraction in web architecture is mainly to supply suitable
subjects for this kind of metadata. Relating information resources to
metadata in this way makes concrete the value proposition for the rule
that a URI should name the information resource related to dereference
of that URI. It is hoped that this analysis will be of use in future
work aimed at strengthening or modifying consensus around this rule.
Received on Wednesday, 18 May 2011 01:05:41 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:48:35 GMT