W3C home > Mailing lists > Public > public-rdfa-wg@w3.org > May 2010

Re: RDFa DOM API Document update

From: Mark Birbeck <mark.birbeck@webbackplane.com>
Date: Thu, 20 May 2010 14:14:39 +0100
Message-ID: <AANLkTimLV7eUar6pvMmS7nkIpSDc6IK-EtmZUbJHB6rJ@mail.gmail.com>
To: Toby Inkster <tai@g5n.co.uk>
Cc: Manu Sporny <msporny@digitalbazaar.com>, RDFa WG <public-rdfa-wg@w3.org>
Hi Toby,

Ah...I see what you mean; when using @rev then it's actually the
subject that triggers triple generation.

Good point, and something we should definitely take into account when
choosing the wording for this.

The essential behaviour we need is that at steps 6, 7, 9 and 10 of the
RDFa core parsing algorithm, at the points where we say "generate a
triple", the parser also needs to store the value of [current
element].

I wonder if the RDFa Core document should cross-reference the RDFa API
at this point, or at the very least flag up that generating a triple
might involve more than simply storing subject, predicate and object.

Regards,

Mark

On Wed, May 19, 2010 at 1:19 PM, Toby Inkster <tai@g5n.co.uk> wrote:
> On Tue, 18 May 2010 23:25:53 +0100
> Mark Birbeck <mark.birbeck@webbackplane.com> wrote:
>
>> Doing this has the effect of keeping track of all of the *objects* in
>> the tree (in the RDF sense of objects, not the OO sense), because it's
>> always objects that cause triples to be generated. (@typeof is not an
>> exception -- it just happens to be an abbreviation for a
>> predicate/object pair.)
>
> Just to be difficult, sometimes the subject causes the triples to be
> generated. e.g.
>
> <div resource="[foaf:Person]">
>  <p property="ex:note">Here are some foaf:Persons.</p>
>  <ul rev="rdf:type">
>    <li src="#alice" property="foaf:name">Alice</li>
>    <li src="#bob" property="foaf:name">Bob</li>
>    <li src="#carol" property="foaf:name">Carol</li>
>  </ul>
> </div>
>
> Though in that particular case, the subject node is probably more
> useful to return!
>
> And to answer Manu's question, yes, this technique does satisfy me. It
> might not always return the best possible node, but at least it's a
> fairly simple and predictable rule. We need to explicitly state it
> somewhere though!
>
> --
> Toby A Inkster
> <mailto:mail@tobyinkster.co.uk>
> <http://tobyinkster.co.uk>
>
>
Received on Thursday, 20 May 2010 13:15:18 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:19:47 UTC