W3C home > Mailing lists > Public > public-sweo-ig@w3.org > January 2007

RE: "RDF/XML Lite" task force?

From: Ian Davis <Ian.Davis@talis.com>
Date: Thu, 11 Jan 2007 13:06:28 -0000
Message-ID: <DD5E887552496241BC701548837A282F0519A7F0@nemo.talis.local>
To: "Benjamin Nowack" <bnowack@appmosphere.com>, "W3C SWEO IG" <public-sweo-ig@w3.org>

I would be very interested in contributing to this effort. I've written
various syntax proposals etc over the years. My current favoured
solution is the "profile of rdf/xml" approach as explained here:

http://iandavis.com/blog/2004/10/constraining-rdfxml

(/me notes a comment from bengee at the end :)

At Talis we have written a serialiser in Java that follows these rules
which we're using to generate RSS 1.0 (for example:
http://api.talis.com/bf/stores/ukbib/items?query=semantic). We currently
use Jena's serialiser elsewhere but we'll switch to this deterministic
algorithm in the coming weeks. We're heavy users of XSLT with RDF/XML
and have a bunch of idioms and tricks to make sure we cater for some of
the variability of the output but making this easier by constraining the
XML output is high on our agenda.

I've said many times that having _any_ XML serialisation of RDF was, in
retrospect, a big mistake because it implied compatibility with the XML
toolset - but it's too late for that so we should work with what we
have. As an aside, I'm also interested in a JSON representation...

Ian


> -----Original Message-----
> From: public-sweo-ig-request@w3.org 
> [mailto:public-sweo-ig-request@w3.org] On Behalf Of Benjamin Nowack
> Sent: 11 January 2007 12:11
> To: W3C SWEO IG
> Subject: "RDF/XML Lite" task force?
> 
> 
> 
> OK, this would probably fit better in SWD, but it's related 
> to SWEO as well. Just some food for more thought:
> 
> One of the (IMO) most destructive SemWeb misconceptions is 
> the "RDF = RDF/XML" one. I guess we all agree that one of 
> SWEO's objectives should be to provide (links to) 
> easy-to-grok material for SemWeb newbies that make the 
> distinction clear. However, if want to go beyond these simple 
> education tasks, we have to reach out and get more people of 
> the larger Web dev community involved. I personally don't 
> have a problem with RDF/XML, and I always sigh when the 
> "RDF/XML sucks" perma-thread reappears every other month, but fact is:
>  * there *is* a perma-thread
>  * it's sucessfully used by SemWeb opponents (and proponents
>    as well, for that matter) to hinder community growth
>  * even many RDFers don't like it
>  * it *is* the recommended syntax
>  * we are not in a position to make everyone switch to
>    e.g. turtle; embedded RDF approaches don't work for
>    all use cases, and generally, an XML-based syntax
>    makes a lot of sense
>  * even the most simple SemWeb "hello world" will include
>    some serialization, any useful *2nd step* will include
>    parsing/consuming
>  * developers didn't forget the pain and frustration when 
>    they gave up on trying to write an RSS 1.0 parser
> 
> So, assuming our wake-up efforts are a huge success and 
> everyone gets interested in SemWeb development, how can we 
> make sure that this 2nd step mentioned above doesn't become a 
> showstopper?
> Of course, turtle is one way, but I think we'd be 
> significantly more successful if we could offer an RDF/XML 
> subset that's easy to write, and (more importantly) easy to 
> parse with existing XML tools. And I believe the main benefit 
> wouldn't be a technical, but a marketing/motivation one. When 
> I wrote my RDF/XML parser in PHP using libxml, I managed to 
> have a basic version running in less than an afternoon, but 
> it took me months before it covered all the different 
> optional features in the spec.
> 
> 
> I'm not sure what my concrete proposal for SWEO would be, I'd
> at least like to see
>  * an as-short-as-possible "essential RDF syntax" document,
>    that, after reading it once, allows developers to write valid
>    RDF/XML (and turtle?) documents.
> 
> What I'm dreaming of is:
>  * a more spec-like "RDF/XML Lite", that is a valid subset
>    of RDF/XML, XSLT and XML parser-friendly, and that allows
>    average coders to easily create conforming parsers and/or
>    converters
>  * we manage to persuade toolkit developers to offer this
>    serialization as an output option
>  * we manage to persuade app/extension/plugin developers to
>    update their RDF/XML export formats
>  * alternatively we manage to deploy some simple
>    rdfxml2rdfxmllite scripts
> 
> I think this would be doable, it's a little bit like DOAP, or
> the foafnet effort from a few years back, just more general.
> I think it would be easy to get consensus on the features this
> RDF/XML subset should (not) have, Leo even started a wiki 
> page[1] some time ago. What's missing is just someone to 
> collect requirements, write it up properly and some authority
> to spread the word.
> 
> Congrats, you reached the end of this post. Fell free to
> tell me this is entirely off-topic ;)
> Ben
> 
> [1] http://esw.w3.org/topic/SimpleRdfXml
> 
> 
>
 
The very latest from Talis
read the latest news at www.talis.com/news 
listen to our podcasts www.talis.com/podcasts 
see us at these events www.talis.com/events 
join the discussion here www.talis.com/forums 
join our developer community www.talis.com/tdn 
and read our blogs www.talis.com/blogs

 
Any views or personal opinions expressed within this email may not be those of Talis Information Ltd. The content of this email message and any files that may be attached are confidential, and for the usage of the intended recipient only. If you are not the intended recipient, then please return this message to the sender and delete it. Any use of this e-mail by an unauthorised recipient is prohibited.
Received on Thursday, 11 January 2007 13:33:25 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:17:35 GMT