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

Re: Interesting experience: my foaf in RDFa

From: Mark Birbeck <mark.birbeck@formsPlayer.com>
Date: Fri, 31 Aug 2007 14:42:48 +0100
Message-ID: <a707f8300708310642h270fe4cbs5d719eb9a32410cc@mail.gmail.com>
To: "Hausenblas, Michael" <michael.hausenblas@joanneum.at>
Cc: "Keith Alexander" <k.j.w.alexander@gmail.com>, "Ralph R. Swick" <swick@w3.org>, "W3C RDFa task force" <public-rdf-in-xhtml-tf@w3.org>

Hi Michael,

I hope you're managing to get some sleep! :)

On 31/08/2007, Hausenblas, Michael <michael.hausenblas@joanneum.at> wrote:
> Regarding the RDFa profile [1], I had a chat with DanC
> recently [2] where he very rightly noticed that
> we have NOT resolved the profile issue for RDFa, yet.
> I'd very much appreciate if we fix this ASAP.
> Ralph, anything I can do in supporting you finalising this?
> Cheers,
>         Michael
> [1] http://www.w3.org/ns/rdfa/
> [2] http://chatlogs.planetrdf.com/swig/2007-08-24.html#T19-54-08

I don't know whether this (the profile attribute issue) falls into the
category of 'agreed on but not voted on', but I do know we've
discussed it in some detail. The last discussion I remember was one
where we agreed that a value for @profile would be supported, but that
it would be optional. (I.e., RDFa might still appear in a document
that does not have a value for @profile.)

The key point is that we don't want to make @profile mandatory,
because that loses us a lot of our use-cases. And since a lot of RDFa
syntax is already present in HTML, it doesn't make a lot of sense
anyway, to insist on a value in @profile. This is because you could
run any HTML document you like through an RDFa processor, and obtain
triples for 'next', 'previous', etc., even if there is no @about, no
@instanceof, no @datatype, etc.

However, if authors want to say, "this document really, really,
really, contains RDFa"--or put another way, the author wants to say,
"this document contains more significant triples than the basic
triples that you might get from an ordinary HTML document"--then our
conclusion was that we should provide them with a mechanism.

DanC's Joost example is a good one; if a publisher that has complete
control over their document formats wants to say 'this document
contains RDFa', they should be able to. But to re-emphasise the main
point, if a publisher who has _no_ control over the profile attribute
in their documents--perhaps someone using a hosting service for their
blog, or adding handbook pages to Drupal, or entering a calendar entry
in Google Calendar, etc., etc.--wants to add RDFa to their mark-up,
why shouldn't they be able to?

So metadata consumers will need to decide for themselves whether they
look for triples in documents that have no RDFa value for @profile. It
certainly would make very little difference for crawlers like Google
to parse anyway, and it equally makes very little difference in
client-side software like Operator, to parse anyway. But of course
there may well be situations where a processor won't bother parsing if
it doesn't find the right @profile value. This would be particularly
appropriate, for example, in situations where the publisher's practice
is known; if I was consuming pages from the Joost server I could
ignore those with no @profile value, if it was well known that they
have taken the 'always use @profile' approach.

By the way, no URI for this has been devised yet, so I guess we need
to sort one out.



  Mark Birbeck, formsPlayer

  mark.birbeck@formsPlayer.com | +44 (0) 20 7689 9232
  http://www.formsPlayer.com | http://internet-apps.blogspot.com

  standards. innovation.
Received on Friday, 31 August 2007 13:42:59 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:50:24 UTC