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

Re: Facebook - RDFa in Open Graph Protocol

From: Melvin Carvalho <melvincarvalho@gmail.com>
Date: Wed, 21 Apr 2010 21:55:21 +0200
Message-ID: <p2h9178f78c1004211255j683701eajd9a22cacdaba803b@mail.gmail.com>
To: Dan Brickley <danbri@danbri.org>
Cc: public-rdfa <public-rdfa@w3.org>, David Recordon <recordond@gmail.com>
2010/4/21 Dan Brickley <danbri@danbri.org>

> On Wed, Apr 21, 2010 at 9:39 PM, Melvin Carvalho
> <melvincarvalho@gmail.com> wrote:
> > 2010/4/21 Dan Brickley <danbri@danbri.org>
> >>
> >> So this just got a high-profile launch at Facebook's F8 conference a
> >> couple hours ago -
> >>
> >> http://opengraphprotocol.org/
> >> http://developers.facebook.com/docs/opengraph
> >>
> >> They're using RDFa, putting structured data into Web sites to sit
> >> alongside a Facebook "Likes" button, so that the topic of the page
> >> (movie, restaurant, book, whatever...) can be understood by apps
> >> downstream dealing with the social data. The RDFa is pretty basic, and
> >> I warned David that there's a good chance they'll be jumped upon by
> >> 100s of well-meaning semweb advocates arguing that they should be
> >> using more existing vocabs, different syntax structures, etc etc. I'd
> >> urge you all to go gently on that front for now, and focus instead
> >> more on how we can improve RDFa tooling and specs than on lobbying for
> >> improvements.
> >
> > One question I have:
> >
> > <html xmlns:og="http://opengraphprotocol.org/schema/">
> > <head>
> > <title>The Rock (1996)</title>
> > <meta property="og:title" content="The Rock" />
> > <meta property="og:type" content="movie" />
> > <meta property="og:url" content="http://www.imdb.com/title/tt0117500/"
> />
> > <meta property="og:image" content="
> http://ia.media-imdb.com/images/rock.jpg"
> > />
> > ...
> > </head>
> > ...
> > </html>
> >
> > should some of the property attributes be rels?
>
> They have chosen to represent URIs as strings, rather than use RDFa's
> built-in representation of URIs. This probably isn't ideal from a
> modelling perspective, and means they're opting out of the ability to
> use relative links in the attribute value. I'd prefer to see typed
> links here but I can understand the appeal of this current highly
> regular notation. I'm not losing sleep over this. I think the biggest
> shortfall is not making more explicit use of typing, but hey, step by
> step!
>

Agree that there are some shortcomings, but on first impressions, this is
looks like a big step forward.  Well done Facebook!


>
> cheers,
>
> Dan
>
Received on Wednesday, 21 April 2010 19:55:50 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:15:06 UTC