UZH Ontologies

[Jonas Tappolet]

Hello baetlers,

 

As some of you might know, we have been developing ontologies at the University of Zurich to be used for Software Analysis. The idea is to interconnect as many Software Engineering data sources as possible to improve the quality of e.g. bug prediction. So far there are 3 ontologies covering Bug Tracker, Versioning System and Source Code where the inspiration comes from Bugzilla, CVS and Java. The ontologies are available under their respective namespaces:

 

http://www.ifi.uzh.ch/ddis/evoont/2008/02/bom/ (Bug Ontology Model)

http://www.ifi.uzh.ch/ddis/evoont/2008/02/vom/ (Version Ontology Model)

http://www.ifi.uzh.ch/ddis/evoont/2008/02/som/ (Software Ontology Model)

 

We should focus on the comparison of baetle and the bug ontology (and partially the version ontology) to end up in a combined version (There’s nothing more annoying and fatal for the SW than multiple ontologies covering the same domain). I will (probably today or tomorrow) create the human readable version of the ontologies aka plain old HTML.

 

So far, looking forward for your feedback

 

Jonas

 

 

________

Jonas Tappolet, Research Assistant, University of Zurich

http://www.ifi.uzh.ch/ddis/people/jonas-tappolet/ | foaf

 

[Erling Wegger Linde]

Hi Jonas,

On Tue, Jun 10, 2008 at 3:14 PM, Jonas Tappolet <tapp...@ifi.uzh.ch> wrote:
> Hello baetlers,
>
>
>
> As some of you might know, we have been developing ontologies at the
> University of Zurich to be used for Software Analysis. The idea is to
> interconnect as many Software Engineering data sources as possible to
> improve the quality of e.g. bug prediction. So far there are 3 ontologies
> covering Bug Tracker, Versioning System and Source Code where the
> inspiration comes from Bugzilla, CVS and Java. The ontologies are available
> under their respective namespaces:
>
>
>
> http://www.ifi.uzh.ch/ddis/evoont/2008/02/bom/ (Bug Ontology Model)
>
> http://www.ifi.uzh.ch/ddis/evoont/2008/02/vom/ (Version Ontology Model)
>
> http://www.ifi.uzh.ch/ddis/evoont/2008/02/som/ (Software Ontology Model)
>
>
>
> We should focus on the comparison of baetle and the bug ontology (and
> partially the version ontology) to end up in a combined version (There's
> nothing more annoying and fatal for the SW than multiple ontologies covering
> the same domain). I will (probably today or tomorrow) create the human
> readable version of the ontologies aka plain old HTML.

I totally agree with you!

I see you are using the Workflow ontology, you should join our
discussion (currently there are several threads on the mailing list)
on whether to use the http://purl.org/vocab/lifecycle/schema# instead
(or in addition?). (see
http://groups.google.com/group/baetle/browse_thread/thread/e25604d7828016d1
and http://groups.google.com/group/baetle/browse_thread/thread/7453c65c5f0b62eb
++).

We have also discussed using properties like isMajorFor with domain
doap:Project see http://code.google.com/p/baetle/wiki/Examples. Which
is probably a better solution as an issue/bug can have different
priorities for each project it might be related to
(directly/indirectly). (see
http://groups.google.com/group/baetle/browse_thread/thread/82e6bd90daa7f0d9).

Well, those are two topics whom we can start ("the merging") with at least :)

- Erling

>
>
>
> So far, looking forward for your feedback
>
>
>
> Jonas
>
>
>
>
>
> ________
>
> Jonas Tappolet, Research Assistant, University of Zurich
>
> http://www.ifi.uzh.ch/ddis/people/jonas-tappolet/ | foaf
>
>
>
> >
>

--
Med vennlig hilsen
Erling Wegger Linde

[Henry Story]

Hi Jonas,

I entered the bom ontology URL into Swoop [1]

It looks like you have some contradiction in your ontology here as
Issue and Defect are shown, once pellet is started, as being
subclasses of the empty set. That may be something to look into :-)

Some thoughts:

I think it would be good to add annotations onto each class and
relation, both tags, so that it should appear more readable in things
like tabulator, and some descriptions to help people browsing the
ontology read it. You could reference chapters in your thesis (with a
url when that is online)

Another advantage is that this will allow you to generate an html
version of your ontology for human browsing the way foaf does it.

http://xmlns.com/foaf/spec/

is generated directly from the ontology and an html template. I used
their python scripts in the AtomOwl ontology which you can find in the
www directory hierarchy of https://sommer.dev.java.net/

Now if the software to do this used jython, or some java vm byte code
library it would be a lot easier to run this script on different
platforms. Well I find setting up python to be somewhat complicated...

Henry

[1] http://code.google.com/p/swoop/