Don’t call me DOM

2 September 2004

Using the Semantic Web for my QA work

Filed under:

My work at W3C is supposedly split into two different activities: as a member of the SysTeam, I’m setting up and implementing projects to ease the operations of W3C work; as a member of the QA Team, I’m trying to work on making W3C end products – mainly, its specifications- “better”, part of the job being to define what “better” means.

Where does my work in Semantic Web projects fit, then?

Well, most of the time (GRDDL being the most notable exception), it fits where it should: as using the right technologies for the right project; that is, since I have been convinced for quite some time now that these technologies help in the day to day operations, I’ve been using them both as part of my pure software development projects, and as part of my QA time – note that this well-defined separation only exists on the paper, and in fact, I like it the best when I work on projects that ease W3C operations and make its specifications better…

For instance, the TR automation project, building on Semantic Web technologies to create a formalized version of the W3C Technical Reports list on which I worked as part of my former Webmaster duties, has allowed me to build a number of tools that makes my job in QA easier, like the TR references checker, or even more importantly, the automation of the QA Matrix maintenance.

(Note that in return, the fact that the matrix data were available in RDF made it easy to integrate information back into the TR page, like the links to the errata; and as time goes, I expect these data will move back and forth again, especially with soon IPP getting into the dance…)

Two small examples of recent hacks that are made easy by having used RDF et al in the first place:

  • a set of N3 Rules allows to quickly detect what expected properties in recently published specifications are missing; for instance, a Candidate Recommendation document should have an implementation report, and if none is known to the system, this will let me know
  • since one of the activities through which we try to improve specifications is to review them, this other set of rules is handy to get the list of recently published First Public Working Draft, which are one of the specific targets of our reviews

This isn’t much; this is no killer app; it’s just damn practical.

Picture of Dominique Hazael-MassieuxDominique Hazaël-Massieux (dom@w3.org) is part of the World Wide Web Consortium (W3C) Staff; his interests cover a number of Web technologies, as well as the usage of open source software in a distributed work environment.