W3C home > Mailing lists > Public > public-rdf-in-xhtml-tf@w3.org > June 2007

Re: [RDFa] ISSUE-2: custom attributes for RDF shorthand

From: Niklas Lindström <lindstream@gmail.com>
Date: Tue, 19 Jun 2007 02:27:54 +0200
Message-ID: <cf8107640706181727t8885ff6y7f4f9149e36968de@mail.gmail.com>
To: "Ben Adida" <ben@adida.net>
Cc: RDFa <public-rdf-in-xhtml-tf@w3.org>, "SWD WG" <public-swd-wg@w3.org>


I agree not to implement it (at all for now). Spontaneously, it seems
to break down the level of abstraction that RDFa is based on and
disrupt the consistency, which would make it harder to learn.

Basically I think the arbitrariness of the value type of the object is
a problem (though I suppose it's based on the RDF/XML equivalent, i.e.
Literals). Also, (qname) attributes should probably not signify
special meaning in and of themselves, unless they belong to a
namespace of special meaning. So I suspect this shorthand is something
of a can of worms. (Though one way of control could be e.g. some
@shorthand-prefixes="dc foaf" attribute directive, but that also
departs from simplicity..)

So in short: no (at least for RDFa 1.0).

Best regards,

On 6/19/07, Ben Adida <ben@adida.net> wrote:
> Hi all,
> I have a "list of easily resolvable issues." We want to resolve these
> ASAP, using email preferably. In other words, please respond ASAP with
> your opinion on this. I encourage you all to consider that we need to
> not change or complicate the XHTML1.1+RDFa specification too much anymore.
> This email concerns ISSUE-2. A separate followup email will be sent for
> each remaining issue. Since ISSUE-2 is not very well documented in the
> tracker, I'll summarize it here:
> there was a proposal to allow for attribute shorthands for certain
> hidden triples, e.g:
> <div about="/foo/bar" dc:title="Foo Bar">
> ...
> </div>
> which would yield
> </foo/bar> dc:title "Foo Bar"
> Mark, jump in if I've described this incorrectly.
> I propose that we *not* implement such a feature in XHTML1.1+RDFa. My
> main argument is for simplicity and not having two ways of doing the
> same thing.
> (I'm even tempted to say that XHTML2+RDFa shouldn't have it, either,
> given that it starts to break away from the whole point of reusing
> rendered data as structure. But that discussion doesn't need to happen
> today.)
> What do you think?
> -Ben
Received on Tuesday, 19 June 2007 00:27:57 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:01:50 UTC