Motivation for contribution to libre software projects instead of free-riding

We have tried and emphasize the strategy that organisations can adopt with respect to the maintainability of the in-house applications based on libre software, in a previous paper. We profess that contribution is in no way mandatory, and free-riding is definitely possible for users of libre software. But at the same time, contribution to upstream projects may be one best way to ensure parts of the maintainance effort can be “outsourced” upstream. Still, it may not be an easy task anyway. For instance, it can be difficult to weight on the decisions of upstream developers and have your modifications integrated in the codebase.
It seems that we’re not the only one to have published on this idea recently. Similar arguments are given by by Kris Ven, Herwig Mannaert and Jan Verelst of University of Antwerp (Belgium), according to the presentation they have made at Calibre’s FOSDEM room : Participation of Independent Software Vendors in free and open source software projects (they also published a paper on the same subject). They argue from the standpoint of ISVs instead of an organisation who wants to use libre software for in-house needs, which was our case. But actually I was not much convinced that this were not two similar situations actually. Even though they may have developped the product as an ISV, that seems to be somehow a very custom-made product, and not one product used by so many different and generic entities. Maybe other cases of ISVs publishing more generic products may be more interesting to add to their point. Still, I’m sure that’s a very good finding anyway. I’ll try to keep an eye on bibliography for such pleas for contribution.

Quality Assurance on Debian : 2 interesting talks at FOSDEM

I’ve noticed two interesting talks made at the FOSDEM recently about QA on the Debian project (well, second one, not especially focused on Debian, but Martin speaks from experience with the Debian project of course), whose slides, I think, are really worth reading :

First one is probably more humoristic, although more technical 😉

In any case, there is surely room for much improvement on systematic Quality Assurance in Libre projects.

Monitoring developper’s activity to draw figures for estimation of effort and developpers productivity

In his presentation “Measuring developer activity. Motivation and some approaches“, our collegue Juan José Amor from Universidad Rey Juan Carlos introduced an attempt to use some new tools to monitor Libre (or not) developper’s activity.

The goal is to be able to establish a good calibration of metrics allowing estimation of developpers activity and productivity. Indeed, developper’s activity counted only in numbers of commits made in CVS, or of lines written in these files committed, may not reflect reality. Sometimes, a commit of one single line may have necessited hours of debug…

I think that this approach is indeed interesting. I hope it can be pursued without too many problems with the privacy of individual developers.

Maybe some new tools like beagle, for instance may be helpful in observing developper’s activity… let’s see what will come out of this research. Maybe I’ll volunteer too for testing them, and estimating what kind of productivity I can measure on myself.

WSL2006 : Extended deadline for CFP for Workshop Free Software in Porto Alegre : 2006/02/27

The deadline for the Call For Papers for the Workshop Free Software (WSL 2006) held in Porto Alegre (Brazil) during FISL 7 has been extended to 27th of february 2006.

Take your chance to meet actors of the largest free software community in South America.

More details and CFP : http://www-inf.int-evry.fr/~olberger/weblog/2006/01/17/cfp-wfs2006-7th-international-workshop-on-free-software-porto-alegre-brazil/

Vidéo de la conf JRES sur ProGET en ligne

La vidéo de la conférence portant sur l’expérience du projet ProGET (“Intégration d’applications logicielles libres pour la réalisation d’une plate-forme de travail collaboratif destinée aux enseignants/chercheurs du GET”) que nous avons donnée aux JRES 2005 est disponible en ligne. Bonne conf 😉