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

RE: Complexity of RDFa

From: Rob Vesse <rav08r@ecs.soton.ac.uk>
Date: Tue, 30 Mar 2010 09:58:04 +0100
To: "'Damian Steer'" <pldms@mac.com>
Cc: <public-rdfa@w3.org>
Message-ID: <EMEW3|30540fd272763eab23110b34d2e2adf3m2Y9wZ06rav08r|ecs.soton.ac.uk|00b801cacfe7$1b9ed6f0$52dc84d0$@soton.ac.uk>
> Hi Rob,
> > Perhaps this may just be a case of not having worked with RDFa for long
> > enough to feel truly comfortable with it - what do other people in the
> > community think?
> I think this covers much of it. You're also being pretty nice about html 
> and css, when I find inspect element / firebug essential wrangling tools.

Yeah I think that is perhaps my main issue with RDFa, it took me months to
come around to RDF/XML and to be honest I still hate it!  I've been messing
about with HTML and CSS for far too many years now so don't have to resort
to inspection tools too often these days

> RDF serialisations always feature a degree of variability, as it is the 
> nature of graph syntaxes. Once you add a few shortcuts to the syntax 
> that increases hugely, but having multiple ways write the same thing 
> isn't necessarily a problem. Try working with n-triples all the time. 
> Syntactic sugar is a good thing.
> (As an aside there are occasions where I pick syntax based on shortcuts. 
> turtle is often more pleasant than rdf/xml, but unless you're working 
> with bnodes it's painful to work with nested structures. rdf/xml is much 
> clearer for hierarchies)

Yes I agree that syntactic sugar is a good thing, I'm ambivalent about
RDF/XML being clearer for hierarchies, I guess it is because of the whole
striping structure but still gets very messy if you have large hierarchies.

> However I do sympathise. Treated purely as an RDF serialisation RDFa is 
> fairly eccentric, and on occasions baffling. Here's a reduction of 
> something I (shamefully) wrote:
>     <p about='#a' xmlns:ex="http://example.com/ns#">
>         <span rev='ex:rev' about='#b' typeof='ex:Type'>
>             <span property='ex:toAttribute' content='In attribute'>
>                 <span property='ex:toText'>In text</span>
>             </span>
>         </span>
>     </p>
> That second line was _not_ a good idea. Having worked with RDFa for a 
> while now I'd never include that much on one element.

Yes that is pretty nasty

> Divorcing RDFa from the host language is misleading, though. The variety 
> of methods to produce the same triples are artifacts of the RDFa use 
> cases. As I mentioned elsewhere we've been using RDFa to type the links 
> between University of Bristol staff and organisations. All it took was 
> the addition of rel="foaf:member" and rev="foaf:member" to links in 
> their respective web pages. Rev is one of those awkward bits of RDFa, 
> but in the right place it makes perfect sense.

I agree that you can't divorce RDFa from the host language, and the
@rel/@rev thing was confusing to me at first especially when you have an
element with both on it though I guess it's one of those things that takes
time to get your head around.

> RDFa will benefit from another trip through the w3c process to mature it.


Rob Vesse

PhD Student
IAM Group
Bay 20, Room 4027, Building 32
Electronics & Computer Science
University of Southampton
SO17 1BJ
Received on Tuesday, 30 March 2010 08:59:06 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:04:45 UTC