DragonFly kernel List (threaded) for 2012-03
Re: GSOC project
On 3/21/2012 15:10, Justin Sherrill wrote:
> The primary goal of Summer of Code is to get a student learning as
> part of a group. If a project's already started by a nonstudent, it's
> fine to work on; it will still benefit the student even if different
> code gets committed.
> In any case, you can edit the page to remove or modify or add projects
> as you see fit.
Yes, it's a wiki, I know.
The point was that GSOC project candidates would benefit from a
collaboration of the DragonFly developers, not individuals contributing
As far as primary goals, if the code isn't usable at the end of the
Summer or at least if there's not a viable plan to get the code suitable
for commit in the post-GSOC phase, then the project was not a success
(in my opinion). Either the project itself was ill-defined and not able
to be reasonably accomplished in which case that's DragonFly's fault, or
the student didn't achieve reasonable milestones. The bar for entry is
that student has a good proposal, a good plan, and hopefully can
demonstrate a strong level of coding skill. While gaining experience of
working in a team is a nice benefit if it happens (seems like most of
these projects are mentor-only interaction), I wouldn't consider that
the main objective.
I would only want to see a student project displace developer work in
the case that a) the developer in question is either the mentor or
agrees to work closely with the student or b) the student has the
credentials that inspire confidence that the project will get completed
and written correctly. Otherwise the most likely outcome is scrapping
the whole thing and starting again which, as I said before, wastes
everyone's time. I'm sure the majority of the students want to see
their code committed ultimately.