Re: Official Response to ISSUE-130 from RDF Web Apps WG

On Mon, Feb 27, 2012 at 1:14 AM, Manu Sporny <msporny@digitalbazaar.com> wrote:
> The core of the clarification that the Working Group felt that they
> should make to the specification is the following: It is the
> responsibility of the Host Language to specify which RDFa attributes are
> allowed on certain elements.

Seems adequate. You may record me as "agree" for the purpose of
disposition of comments for this decision.

> This has always been the case for RDFa, but
> the specification text seems to be unclear of this fact in certain
> places. The Editor of the RDFa Core specification has made this clear in
> the latest draft of the specification:
>
> http://www.w3.org/2010/02/rdfa/drafts/2012/ED-rdfa-core-20120223/#hostlangconf

It seems the spec text manages to carry out the decision, though the
key text is "(optional)" next to href and src in section "Attributes
and Syntax".

> Since @href, @rel and @rev were always defined on all elements in
> XHTML1+RDFa, changing this would result in a backwards incompatible
> change and so the Working Group decided to not change this behavior in
> XHTML1+RDFa 1.1.

Please record me as "disagree" for this decision for disposition of
comments purposes. (I disagree that the document conformance
definition for Foo+RDFa 1.1 needs to keep all Foo+RDFa 1.0 content
conforming. Note that conformance is different from the processing
rules.)

> For HTML+RDFa, it was decided that it would be unwise to deviate from
> where some of the more popular attributes, like @href and @src, could be
> placed. The Working Group decided to not override where @href and @src
> are allowed for HTML5 and XHTML5 - expect this change in the next
> version of the HTML+RDFa specification.

I agree with the decision but I can't assess how it's carried out in
spec text before it's carried out.

> Finally, the use of @rel and @rev everywhere cannot be removed without
> cutting two of the more useful features of RDFa - namely forward
> chaining and reverse chaining. Doing so would unnecessarily limit the
> flexibility of the language. So, the Working Group decided that @rel and
> @rev should still be allowed everywhere in HTML+RDFa.

Please record me as "disagree" for this decision for disposition of
comments purposes.

> For the purposes of the W3C Process, all of the resolutions that applied
> to RDFa Core and XHTML+RDFa, resulted in non-substantive changes because
> they were either vagueness or bugs in the specifications.

I disagree with the notion that fixes to substantive bugs don't
constitute substantive changes. I object to recording the changes here
as non-substantive in the disposition of comments.

-- 
Henri Sivonen
hsivonen@iki.fi
http://hsivonen.iki.fi/

Received on Thursday, 15 March 2012 12:47:48 UTC