I have just made some updates to my Semantic Weblogs page with a few new people, mostly what I’ve called occasional semweb bloggers. The ones in the main list are those that feed Planet RDF directly as already described by Edd in Planet Blog but it might not be quite clear how the content gets there.

It starts off with me editing the page above, adding or removing feeds. That page is written in XHTML since it is markup for humans and also a web page. The page is validated as legal XHTML by emacs nxml mode of course. For machines, I add some hints. Once I add an entry, it’s marked up with a <span> and class that describe if the entry is a person or a group which affects the output RDF/FOAF class. This is then processed by XSLT into RDF/XML, the bloggers.rdf file – also known as a FOAF blogroll – with a list of the bloggers, their names, the titles of their blogs (if it has one), their RSS 1.0 feeds etc. I expect this process possibly could be GRDDLed. (Most of the other Planet sites use this FOAF blogroll format that we created).

The bloggers.rdf page is read by the Planet RDF software to seed and control the RSS aggregation. The output of this is the aggregated RSS 1.0 feed (i.e. RDF) which also contains descriptive content from the bloggers.rdf file that is suffiicent for another XSLT script to process it to give the output Planet RDF XHTML web page – both the RSS content in the main content of the page along with the list of blogs in the sidebar.

So, the new blogs added are (in no particular order):

• Binary Relations by Morten Frederiksen
• overflow by dylanb (occasional)
• James Tauber (occasional)
• Tom Hoffman and his sem_web category (occasional)
• Mark Nottingham (occasional)
• Christopher Schmidt (occasional)
• Better Living Through Software by Joshua Allen (occasional)
• PeoplesDNS blog (corporate blog)

What this would benefit from is more RSS 1.0 feeds for categories of blogs; some people already do that for Planet RDF, I can see this being an increasing trend. Planet RDF doesn’t read any RSS Tag Soup feeds at present due to the myth of RSS compatibility. Maybe Atom content will be added later if they can work out what they are identifying and it has a standard RDF mapping. RSS 1.0 works just fine till then and you can’t beat it being stable for nearly 4 years.

Some interesting recent posts I saw from the above include:

• XML Infoset and XML Schemas versus RDF and RDF Schemas by James Tauber
• RDFLib in zOPE via Tom Hoffman
• sparta.py by Mark Nottingham also commented on in Naked Object in Sparta by James Tauber
• Why RDF? by Joshua Allen on Microsoft’s triple-y WinFS and RDF

Jasper van de Gronde popped up on the redland-dev list with some bugs and fixes for Raptor on win32 with MSVC6 (i.e. not cygwin). Turns out he’s been working on what I’ve called RDFDB/win32 which is a much modified port of RDFDB to Win32 and has now switched to using Raptor for the parsing. Somewhat of a similar space to Redland but with a server and not modular internally – hey a competitor :). Jasper has been making this for his MDDB system which is described as a personal MetaData DataBase (hence MDDB) intended primarily for music metadata it seems.

The Courtship of Atom by Kendall Grant Clark (XML.com) reports on the W3C/Atom discussions – would the W3C be a good home for it? I attended the 2004-05-18 ad-hoc meeting in New York that Kendall mentions. I don’t think the Atom people necessarilly see the semantic web activity as the natural home, but it might be the most appropriate at this time. A new W3C Activity is a big deal. My primary interest is in the Atom API, since RSS1.0 is fine for my data needs. The API might link nicely to the RDF Data Access WG work.