- From: Dan Connolly <connolly@w3.org>
- Date: Tue, 19 Jun 2007 05:58:48 -0500
- To: Jeremy Carroll <jjc@hpl.hp.com>
- Cc: GRDDL Working Group <public-grddl-wg@w3.org>
On Thu, 2007-05-24 at 14:48 +0100, Jeremy Carroll wrote:
>
> Summary:
>
> Much better than when I last looked, still a little bit of tidy up
> needed before PR.
>
> =============
>
> This is a review of:
>
> http://www.w3.org/2003/g/data-view
> $Revision: 1.41 $ of $Date: 2007/05/02 13:33:23 $
> by $Author: connolly $
OK, I considered those suggestions and implemented
them, or something close, in
revision 1.44 of 2007/06/19 10:46:59
Details...
> 1)
> The document should link to the specification in the first para.
Done.
> 2)
> Before PR the Note:
> [[
> Note: While we have a growing
> body of documentation, examples, code,
> and experience with GRDDL, it is still experimental and subject
> to changes. The GRDDL Working
> Group welcomes comments
> by email.
> ]]
> should be rewritten.
Done: "Note: As of May 2007, the GRDDL specification is a W3C
Candidate Recommendation. The GRDDL Working
Group welcomes comments
by email, especially comments that reflect implementation
experience."
> 3)
> Suggest the heading:
> [[
> Learning More
> ]]
> should be
> [[
> Definition of the GRDDL Namespace and Metadata Profile
> ]]
Changed to
GRDDL Namespace and Metadata Profile Reference
> 4) in the first reference
> Suggest delete "2
> March 2007"
> the link is an undated URL and so the text should not have a date either.
Well, I think it's useful to refer to it using a non-restrictive
clause a la "the GRDDL spec, which was released May 2007".
When a new version is released, the namespace doc should be
updated, but if it's not, the failure mode is that the reader
gets the newer update, though mismatch in the metadata
suggests that the namespace document is a bit untrustworthy,
which is a reasonable suggestion in that case.
> The undated version seems appropriate.
> (On looking at the GRDDL result, I am unsure about this comment.
> Overall, the conventions on dated versus undated versions are not
> entirely clear).
I edited the citations for consistency:
</TR/grddl/> dc:date "2007-05-02";
dc:title """Gleaning
Resource Descriptions from Dialects of Languages
(GRDDL)""";
:label """Gleaning
Resource Descriptions from Dialects of Languages
(GRDDL), W3C Candidate Recommendation 2 May 2007""" .
</TR/rdf-concepts/> dc:date "2004-02-10";
dc:title """Resource
Description Framework (RDF): Concepts and Abstract
Syntax""";
:label """Resource
Description Framework (RDF): Concepts and Abstract
Syntax, W3C Recommendation 10 February 2004""" .
</TR/webarch/> dc:date "2004-12-15";
dc:title """Architecture of the World Wide Web, Volume
One""" .
</TR/xslt> dc:date "1999-11-16";
dc:title """XSL Transformations (XSLT) Version
1.0""";
:label """XSL Transformations (XSLT) Version
1.0 W3C Recommendation 16 November 1999""" .
> 5) Before PR, I suggest deleting the ToDo item at the bottom of the
> page, either by doing it, or by not doing it.
Deleted (by not doing it).
> 6) Reading the GRDDL namespace with the Jena reader ...
> (this looks OK, but perhaps should be reviewed in detail)
I read it again with the online GRDDL service
and ran it thru cwm's N3 pretty-printer; it looks OK
to me.
I'm also trying to look at it in the tabulator, but
I think the tabulator is kinda unstable/buggy just now.
--
Dan Connolly, W3C http://www.w3.org/People/Connolly/
Received on Tuesday, 19 June 2007 10:59:28 UTC