Glimmers of RDF/XML support

[Ben Campbell]

Hi all (but mostly, I suspect, Dan ;-)

I've made my first small steps at generating RDF/XML versions of
journalist pages to journalisted.

I'm still getting my head around a lot of the concepts, so I fully
expect my implementation as it stands to be torn to shreds :-)

So far I just just output basic details (using FOAF) and some employment
and education information (using DOAC).

eg:
http://journalisted.com/fran-abrams?fmt=rdfxml

Still lots of other things to add - I want to include all the data
already on the html version of the page in the RDF version... just a
matter of finding the right vocabularies to use (I know they're out
there somewhere :-)

Feedback welcome!

Once I'm a bit further along, I'll put a proper URL scheme in place. My
rough plan for that is:

/journo/fred-bloggs - the journo (redirects to HTML or RDF depending on
HTTP "accepts" headers)
/fred-bloggs - html version
/data/fred-bloggs - RDF/XML (or /fred-bloggs.rdf ?)

(and I'll probably aim to support other RDF serialisation formats too)

Thanks,
Ben.

--
You received this message because you are subscribed to the Google Groups "jl-dev" group.
To post to this group, send email to jl-...@googlegroups.com.
To unsubscribe from this group, send email to jl-dev+un...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/jl-dev?hl=en.

[Ben Campbell]

Hi Dan - thanks for the great feedback!
I'll write up a proper reply to it all soon and fix up the worst of the
atrocities. I'm just a bit snowed under by other stuff for a few more
days :-(
Also - yes - I'll set up a database dump for you and write up some
proper notes on setting up a sandbox (I need to scrub out article text
and the user accounts from the database, for legal and privacy reasons)


I'm treating the RDF output as just another view onto the data, on par
with the html version. The collected data is briefly covered in comments
at the top of the HTML template I use:

http://github.com/bcampbell/journalisted/blob/journo-login/jl/templates/journo.tpl.php

The xml/rdf template uses exactly the same set of data:

http://github.com/bcampbell/journalisted/blob/journo-login/jl/templates/journo_rdfxml.tpl.php

(although I'll probably replace this using some library that can
serialise a bunch of different RDF representations - seems silly not to
support that)

There are quite a few things not yet output by the RDF version. Any
suggestions for how to represent missing things are welcome :-)
For the list of articles a journalist has written, it occurred to me
that using RSS 1.0 as a vocabulary might be appropriate? Is this a
totally insane idea?

Once it's all looking a bit more respectable, I should open it up to
wider criticism, rather than just relying on you to do all the work :-)
Are there any particular mailing lists or forums I should look into?