W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > October 2002

Re: Minutes 25 May (2001) for review (and: fun with RDF for meeting records)

From: Jeremy Carroll <jjc@hplb.hpl.hp.com>
Date: Fri, 25 Oct 2002 15:56:14 +0200
To: <w3c-rdfcore-wg@w3.org>
Message-ID: <BHEGLCKMOHGLGNOKPGHDEEHOCAAA.jjc@hpl.hp.com>


I wasn't present (in the group) for this meeting, but it seems to be the one
where it was decided that unqualified attributes from "the List" were not in
RDF.

Since we are being asked to review that decision, I thought I would review
the minutes.

A particular issue that I was trying to get my head round is why DaveB's
preference for deprecating was rejected.

Dave:
[[[
3.  Unprefixed attributes are deprecated - they SHOULD NOT be used in
    the syntax from this date and WILL be forbidden in the next RDF
    syntax document.
]]]
http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001May/0166.html

The minutes record:
[[[
Connolly: licensing processors to accept documents that aren't in the
language is risky... the user community learns what the language is by what
the tools do
]]]

This is a good argument, but we have seen in practice a reluctance on the
part of the tool builders to do more than issue a warning on these
constructs. The tools have effectively deprecated unprefixed attributes.

The warning mode also, to some extent, counters DanC's arguments that "the
user community learns".

So, in reconsidering this issue, my preference would be to reconsider only
part 3 of DaveB's original proposal, and suggest that the syntax doc
specifically note that deprecation warnings on this construct are
appropriate.


===

Being an outsider at the time, I found the original decision surprising.
Reviewing the minutes and the e-mail log I think the arguments for it were,
overall, compelling; and we should not backpedal too far or too fast.


Jeremy
Received on Friday, 25 October 2002 09:56:15 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:52:31 EDT