thanks a lot for making the feed to SIOC XSLT available! I've checked in a version of the feed-sioc.xsl script [1] that generates valid RDF/XML.
One thing: currently, the URL to the feed has to be passed as a parameter to the XSLT. Is it possible to get that information out of the feed itself? I don't have the feed/Forum URL at hand when invoking the script.
The code to extract feed's URL from the link rel="self" element was already there ( thanks, Morten! ), just needed to loosen its matching rules because it was looking for a link of type "application/atom+xml" but there are many feeds that advertise their url as other MIME types such as "text/xml".
Uldis
On 11/28/06, Andreas Harth <andreas.ha...@deri.org> wrote:
> thanks a lot for making the feed to SIOC XSLT available! I've checked > in a version of the feed-sioc.xsl script [1] that generates valid RDF/XML.
> One thing: currently, the URL to the feed has to be passed as a parameter > to the XSLT. Is it possible to get that information out of the feed > itself? I don't have the feed/Forum URL at hand when invoking the script.
On Tue, Nov 28, 2006 at 02:48:52PM +0000, Uldis Bojars wrote:
> Fixed that.
> The code to extract feed's URL from the link rel="self" element was > already there ( thanks, Morten! ), just needed to loosen its matching > rules because it was looking for a link of type "application/atom+xml" > but there are many feeds that advertise their url as other MIME types > such as "text/xml".
I have made some improvements to the XSLT [1] - now categories are handled a bit better, and the XSLT also converts from RSS 1.0 to SIOC.
> I have made some improvements to the XSLT [1] - now categories are handled a > bit better, and the XSLT also converts from RSS 1.0 to SIOC.
I changed XSLT script to output all RDF as-is.
This shall fix the problem that script output was empty if input was RDF other than RSS 1.0, e.g., LJ FOAF. A side effect is that RSS 1.0 will also be output as-is for now.
One small difference in output is that if the source had CDATA sections (e.g., content:encoded) those will be output entity encoded instead.
To All: we are using this script to convert a large amount of feeds in different formats (mainly Atom and RSS 2.0) to RDF and are using SIOC as a universal representation that we can map them all to ("one ring to rule them all").
I have recently started working with SIOC and I am very interested in converting Feeds to SIOC. My name is Luis and I am working on a Semantic enhanced blog search engine for my master Thesis. My master Thesis coordinator is Juan Miguel Gomez who sent you (Uldis) an email and were working for deri before coming back to Madrid.
I am trying to read as much as possible about SIOC, Ontologies, SPARQL... and I can see how much work you have spent on it.
I tried to download your XSLT and it doesnt work? is the URL bad?
Thanks in advance
luis.
On 14 dic, 19:23, "Uldis Bojars" <capts...@gmail.com> wrote:
> > I have made some improvements to the XSLT [1] - now categories are handled a > > bit better, and the XSLT also converts from RSS 1.0 to SIOC.I changed XSLT script to output all RDF as-is.
> This shall fix the problem that script output was empty if input was > RDF other than RSS 1.0, e.g., LJ FOAF. A side effect is that RSS 1.0 > will also be output as-is for now.
> One small difference in output is that if the source had CDATA sections > (e.g., content:encoded) those will be output entity encoded instead.
> To All: we are using this script to convert a large amount of feeds in > different formats (mainly Atom and RSS 2.0) to RDF and are using SIOC > as a universal representation that we can map them all to ("one ring to > rule them all").
Welcome and thanks for spotting mistake in the URL of XSLT script. The URL I posted above should work ok.
> I have recently started working with SIOC and I am very interested in > converting Feeds to SIOC. My name is Luis and I am working on a > Semantic enhanced blog search engine for my master Thesis. My master > Thesis coordinator is Juan Miguel Gomez who sent you (Uldis) an email > and were working for deri before coming back to Madrid.
You have picked an interesting topic. I received Juan's email that you mention, sorry if I have been slow with replies, this has been a very busy time. Welcome to ask anything that you need to know and best copy your questions to the list so that others have an opportunity to reply.
SIOC project site [ http://sioc-project.org ] should be a good introduction. We are also working on more tools that work with SIOC and, as discussed on SIOC-Dev earlier on, applications that can consume SIOC data are what we need now.
> I am trying to read as much as possible about SIOC, Ontologies, > SPARQL... and I can see how much work you have spent on it.
Thanks. There is a solid basis now, but there's much more work that can be done.
In terms of work to be done - data export part is built for WordPress, Drupal and some other CMS engines, but even then more SIOC plugins/exporters need to be created. If anyone know of a CMS / blog / bulletin board engine that 'd benefit from SIOC export, please go ahead and create SIOC plugin for it. There is a SIOC API for PHP that you may use.
We have got a development server now where we shall be able to experiment with SIOC w/o impact on main SIOC site and services. A SIOC search engine would be one thing I'd like to see working. If you have more ideas about applications to build for SIOC, please write.
