let's get ready for Google Summer of Code 2014

[Sebastian Kügler]

On Tuesday, February 11, 2014 11:03:39 Myriam Schweingruber wrote:
> On Tue, Feb 11, 2014 at 10:56 AM, Sebastian Kügler <sebas at kde.org> wrote:
> > On Monday, February 10, 2014 16:21:34 Mark Gaiser wrote:
> >> On Mon, Feb 10, 2014 at 4:13 PM, Vishesh Handa <me at vhanda.in> wrote:
> >> 
> >> No, certainly not. I know "a bit" about KIO, but others know _way_
> >> more. And guiding a student requires someone with more in depth
> >> knowledge then i have. (looking at David Faure ^_-)
> > 
> > That's a misconception. What a mentor has to do is to show a developer the
> > ropes, get him or her started, but the mentor doesn't have to be an
> > absolute authority in a given domain (but has to realize that). As long
> > as you can point the student towards finding a solution, or tell him/her
> > where to find help, whom to ask, it's perfectly fine to mentor someone.
> > 
> > Most of the work in guiding the student is in ... well ... guiding the
> > student, and that's more related to processes such as where to find help,
> > how to get the build set up. In my experience, it's mostly about
> > processes than about absolute knowledge and expertise of a given piece of
> > code.
> 
> That is all very well, but the fact is: we should NOT add project
> suggestions without having a mentor for it, this gives a very wrong
> signal, as students might be interested to work on something but will
> not be able to do so without a mentor. So if you add a project, please
> make sure you have a mentor for it.

The point was to encourage Mark to take up mentoring. Possibly David might be 
a better person to mentor, but in case he's limited in time (and I'm quite 
sure he is), someone else can mentor, and David jumps in when and if the code 
needs a second round of eyeballing.

> And no, it is not enough to be able to guide, you also need people
> available who can review code. I was able to mentor twice in the past
> because I was around to help the student find solutions and nag them
> to give the requested regular feedback and to make sure they respected
> the deadlines and had somebody to talk to when there are problems, and
> I had several people at hand who could do the code reviewing part,
> something I definitely can't do myself. So if nobody is around to
> review the code, then don't suggest a project, as it is a necessary
> step in the process.

I agree, though in practice, the work can often be spread out. What works is 
having one person to do most of the mentoring tasks, and someone else (or more 
people) who can help with reviewing code. Not everything has to be done by a 
single person.
-- 
sebas

http://www.kde.org | http://vizZzion.org | GPG Key ID: 9119 0EF9