Using standardised ontologies for querying bug status from bugtrackers

[Olivier Berger]

Hi.

I'd like to share some thoughts with you, on some future developments
I'm considering for bts-link.

As you probably know, as part of our work on the Helios project, we are
thinking of using bts-link outside its initial scope (monitoring Debian
bugs forwarded to upstream projects' bugtrackers), to monitor relations
between any "local bug" with any other remote bugs, in other contexts.

For doing so, we are trying to encourage bugtrackers to offer a standard
way to describe (semantically) their bugs, using standard ontologies.
Instead of querying the bugtrackers through some web-scrapping custom
adapter, tools like bts-link could then just access the RDF (or similar)
facets of the bugs, and directly retrieve the bug's properties with
non-ambiguous meanings.

We've started demonstrating such approach through 2 prototypes of RDF
exporters of bug properties, based on UDD data about Debian bugs, and on
a bugzilla database (see :
http://www-public.it-sudparis.eu/~berger_o/weblog/tag/rdf/ for links
with more details).

The ontology we're working on
(https://picoforge.int-evry.fr/cgi-bin/twiki/view/Helios_wp3/Web/HeliosBtOntology) integrates properties like helios_bt#reportedUpstreamIn which corresponds to the "forwarded upstream" debbugs attribute.

In the future, I plan to try and adapt bts-link to prepare it to use
such RDF descriptions of bugs when they'll become available, so that
it's ready to explore links between these bugs on the semantic web.

Any comments much welcome, of course.

Best regards,
-- 
Olivier BERGER <olivier.berger at it-sudparis.eu>
http://www-public.it-sudparis.eu/~berger_o/ - OpenPGP-Id: 1024D/6B829EEC
Ingénieur Recherche - Dept INF
Institut TELECOM, SudParis (http://www.it-sudparis.eu/), Evry (France)