Re: [Fwd: [uf-discuss] Fwd: [whatwg] Removing @rev]

Hi!

Well, as long as this doesn't mean @rev may be dropped from (future
domains of) RDFa? I mean, if HTML5 "wins", and e.g. XHTML2
applications will be scarce to non-existant (and when XHTML 1.1(+RDFa)
"eventually" becomes a legacy format), having no support for @rev may
have a bad effect for lots of use cases.

It depends on other things, but since AFAIK @rev reduces the need for
repetition in a lot of cases, such as describing
memberships/composition (e.g. as I describe in
<http://lists.w3.org/Archives/Public/public-rdf-in-xhtml-tf/2008Jan/0091.html>),
it (probably) has a direct effect on the ease of building end-user
applications that produce RDFa (since e.g. repetition of uri:s
requires parallell maintenance to handle the indent).

[Digression: It's a complex issue of course, but I'm uncomfortable
with the way HTML5 seems to drop stuff that are relevant to many good
things, e.g. RDFa. Well, @profile and @rev to date(?), but who knows
what else may be thrown out since "nobody uses it".. I hope for a
future for XHTML2. (Or even something like "web document markup" and
an extension like "web application markup", but that's another
story.)]

Just my two cents.

Best regards,
Niklas



On Fri, May 16, 2008 at 4:27 AM, Ben Adida <ben@adida.net> wrote:
>
> Manu Sporny wrote:
>>
>> This has bearing on RDFa in HTML5... I had heard that @rev was being
>> preserved from somewhere as well... looks like it's on the way out.
>
> I noticed this while building my cleanroom RDFa parser on top of the HTML5
> parser. It actually strips @rev out. I don't understand why it goes to that
> extent when other attributes, e.g. "foobar", make it into the parse tree
> just fine. In any case, RDFa can function fine without @rev. I think it's
> rather pointless to get rid of it just because it supposedly confused some
> authors (those people can just keep on ignoring it), but if that's the
> decision, it's no significant skin off our backs.
>
> (Side note: my only delay in the cleanroom implementation is that all of the
> SPARQL engines require RDF/XML and not N3... currently passing 70-80% of the
> test cases...)
>
> -Ben
>
>
>
>

Received on Friday, 16 May 2008 16:26:42 UTC