[Soc-coordination] Status of the Taverna/BOINC/Wrapper/GeneticCloning projects

[Obey Arthur Liu]

Hi all,

On Wed, Apr 6, 2011 at 2:49 PM, Ana Guerrero <ana at debian.org> wrote:

>
>  {*} [[SummerOfCode2011/JavaPackages]]: Java packages for Taverna
>
> The project does not have any public student proposal. It has a mentor.
> The project seems to fail in the "package this category" mentioned in this
> mail [0].
>
>
>  {*} [[SummerOfCode2011/BOINC]]: Streamline the preparation of BOINC
> projects
>
> The project has 2 students proposal in melange and 1 mentor commited
> partially
> to at least another 2 more project.
> The discussion and comments in melange point that the purpose of the
> project
> and its deliverables are not clear.
>
>
>  {*} [[SummerOfCode2011/Wrapper]]: A wrapper aware of dependencies for
> direct
>  remote execution
>
> The project has 1 students proposal in melange and 1 mentor and 1
> co-mentor.
> The discussion and comments in melange point that the purpose of the
> project and its deliverables are not clear.
>
>
>  {*} [[SummerOfCode2011/GeneticCloning]]: Genetic Cloning with Debian
> The project has 2 students proposal in melange and 1 mentor commited
> partially
> to at least another 2 more project.
> The discussion and comments in melange point that the purpose of the
> project,
> its deliverables and how the project benefits to Debian is not clear.
>
>
> [0] http://lists.debian.org/debian-devel/2011/03/msg00972.html
>
>
As a long time admin for the Google Summer of Code at Debian, I feel that I
need to chime in on the thorny issues with the projects quoted above.
This might not be popular with everyone but they should be removed from the
Google Summer of Code at Debian immediately.
Let me explain why.

- These projects do not further the technical objective of contributing to
Debian as a primary concern.
I feel that the technologies involved and contributions expected from and
for Debian are secondary to the scientific aspects. Getting code written is
the endeavor of the Summer of Code but the goal of the Summer of Code at
Debian is to specifically contribute to Debian. This and only this is our
mandate.

- These projects are not targeting teams that span the enough core parts of
Debian community.
The GSoC at Debian has a very specific objectives of introducing students to
broad-ranging parts of the Debian community with the objectives of teaching
them what Debian is and giving them a rather unique opportunity to engage
themselves as Debian Developers in the long run.

- It should be clear that Debian as an organization is about creating an
operating system distribution. Debian (the organization) is not a scientific
organization. I feel that the core interests of most students applying for
these projects would be much better served by applying at actual scientific
organizations (of which there are many participating in GSoC)

- In relation to the previous point, the realistic evaluation points that
will be used for these projects (successfully running specific research
experiments) are too removed and intangible for the Debian development
community at large to satisfy our requirements for transparency,
accountability and reusability.

- Debian has no obligation to fulfill all the slots Google may give. It is
in no one's interest to manage more projects than should be. Each project is
an additional management burden that should be decided on on an individual
basis, weighted with the expected outcomes.

As a additional note, I should point out that the only reasons these
projects have stayed up for so long is that they were mentor and
DD-introduced. Similar projects submitted by students the previous years
have been immediately notified of the various issues suggested above and
forwarded to more relevant organizations. DDs should be held to the same
standards.

Lastly, I would like to apologize to the impacted students for being caught
in this unfortunate situation, and encourage them to still seek a Summer of
Code project, which is a wonderful experience.

Regards,

-Arthur
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/soc-coordination/attachments/20110406/915804a0/attachment.htm>