Re: Google Summer of Code 2008
On Wed, Feb 27, 2008 at 7:11 PM, Lucas Nussbaum
<lucas@lucas-nussbaum.net> wrote:
> On 27/02/08 at 16:33 +0100, Ondrej Certik wrote:
> > If some projects in the past were a failure, it is solely the problem
> > of the management (=student's mentor:), it doesn't matter if the
> > student was or wasn't a DD. If the student is working on something
> > else (doesn't matter it is also related to Debian), his mentor should
> > fail him in the middle summer evaluation.
>
> What if the student worked a bit on his project, but couldn't work a lot
> because he was too busy working on a huge transition in Debian, or on
> organizing some Debian conference? You realize that such situations
> will never be black and white, and that failing a student (who might be
> a friend, or at least a fellow DD you had some beers with) is a very
> hard decision to take for a mentor?
Absolutely, I agree it is not black and white. But this is the
responsibility of the mentor. He needs to be the one, who makes this
decision and he needs to stand behind this decision.
So for example if the student failed to do his GSoC project, but he
organized a Debian conference and/or other things, this is what should
imho be done:
* the mentor should decide if he should fail him or not
* at the end of the GSoC, the student (and/or mentor) should write a
sum up, and this should be in the wiki. So that google and other
people can see it and see for themselves, whether the $4500 from
Google were justifyably spent on this project.
So to answer your question - yes, I think it is perfectly possible not
to do every single bit of the application, but in this case, the
mentor needs to make sure he can back up his decision not to fail the
student. Publicly.
That's all I ask. As the outsider, all I see is this half empty wiki:
http://wiki.debian.org/SummerOfCode2007
Instead, there should be reasons why each student either succeeded, or
failed. Or at least some summary what each student did.
The reason why I am saying this is that you are afraid of Google
giving less slots each year, because the projects were a failure.
> So there won't be a shortage of candidates, even if we decide to forbid
> curent contributors from applying.
Probably. But I don't think that restrictions of applicants can help
achieve what you want (i.e. projects not being a failure). I believe
in the exact opposite - allow everyone to apply, choose the best and
require to fullfil the commitments as stated by the students
themselves in the application. Plus showing progress of all students
publicly.
Ondrej
Reply to: