W3C home > Mailing lists > Public > semantic-web@w3.org > April 2010

Re: POWDER support in Perl (was Re: Hacking RDF with Perl hackathon cancelled)

From: Phil Archer <phil@philarcher.org>
Date: Mon, 19 Apr 2010 09:16:42 +0100
Message-ID: <4BCC116A.7020701@philarcher.org>
To: Kjetil Kjernsmo <kjetil@kjernsmo.net>
CC: dev@lists.perlrdf.org, Dan Brickley <danbri@danbri.org>, Semantic Web <semantic-web@w3.org>
Hi Kjetil,

You are right in every particular. The POWDER processor I built is 
written in Perl and uses a couple of off-the-shelf modules (XML::Parser, 
XML::Parser::PerlSAX being the main ones to parse a POWDER file which is 
then split into various MySQL tables).

What I have not done is to tidy up the code to make it transportable. 
That's one of the reasons for not having made the code available as a 
download to date. The other reason is that I'm a big girl's blouse and 
I'm far from confident that my code isn't just too darned amateur to be 
exposed to the likes of serious Perl folk who would have to decide 
whether to laugh or cry when they see just how ugly it all is! Having 
said that, what the heck...

I've now zipped all the Perl code up into an archive that can be 
accessed via http://i-sieve.com/powder/ which is where I put all the 
i-sieve stuff about POWDER (there's a link at the bottom of that page to 
http://i-sieve.com/powder/code.zip). This is just a cope of the working 
code with a few internal file system references removed for security 
(oh, and I remembered to delete the MySQL password ;-) ).

I have not done any work on this for a while. I'd love to, but both 
i-sieve and W3C pay me to do other things, not POWDER, so I don't have 
the bandwidth to spend on this sadly.

If I had time (and skill) I would develop a POWDER system that:
- Implemented the full POWDER -> POWDER-BASE -> POWDER-S transformations 
(the former is already available, the latter requires an XSLT engine. To 
do the full transformation in one XSLT requires XSLT 2.0.)

- Build a SPARQL interface so that you could run a query directly 
against a POWDER doc.

- Add to the processor so that it could populate a triple store as 
managed through Redland. The output of the processor is RDF triples but 
there's no triple store there as such. And, Schrödinger's cat-like, the 
triples only come into existence when you ask for them so there's a bit 
of engineering to be done there.

Thanks for raising the issue,



Kjetil Kjernsmo wrote:
> Just a quick comment on:
> On Sunday 18. April 2010 21:48:39 Dan Brickley wrote:
> Phil Archer has a processor for POWDER, which written in Perl but to my 
> knowledge is only online as a CGI script:
> http://i-sieve.com/cgi-bin/processor.cgi
> Phil, could you please comment?
> Toby gave a good overview of the Perl support for RDF, I'll try to find 
> time to respond as well.
> Cheers,
> Kjetil


Phil Archer
+44 (0)1473 434770

i-sieve Technologies                   |      W3C
Sentiment Analysis Beyond Impressions  |      Open Media Web
http://i-sieve.com                     |      http://www.w3.org
Received on Monday, 19 April 2010 08:17:27 UTC

This archive was generated by hypermail 2.4.0 : Tuesday, 5 July 2022 08:45:17 UTC