[Kde-games-devel] Re: GSoC and Knights

[Parker Coates]

On Mon, Mar 14, 2011 at 16:50, Miha Čančula wrote:
> Hello!
>
> As you might know, I'm the author of the KDE4 version of Knights. The
> program has just released a stable 2.3 version with enough features for
> general use, but there are still things missing. I posted a GSoC idea on the
> Techbase page and got some responses from students. I have another idea in
> mind, but I won't publish it on the techbase if there's no reason to.
>
> Unfortunately, I myself am still a student, and I would like to participate
> in the summer of code program as well. I can't be both a student and a
> mentor at the same time, and considering that I could use some extra money,
> I'd rather be a student. However, I've never participated in GSoC (but I
> have in SoK), so I can't say if I'll be accepted or not. This brings us to a
> dilemma, and I'd like some things cleared first.

I'm guessing that getting a GSoC placement in your own project with an
outside mentor isn't really allowed?

> 1. Would a proposal for Knights be accepted at all? I know that you're
> reluctant to accept proposals for new games, but this project would not be a
> new game. Then again, it's part of extragear and not the kde-games module,
> and as the first game there, I have no idea what the situation is like.

Although Knights may not technically be in the KDEGames module, it is
definitely a KDE application and a member of the KDE community. I
don't see any reason why it wouldn't be accepted, but it might be
worth confirming with the KDE GSoC folks. After all, Amarok is /only/
an extragear app and it gets plenty of GSoC attention. ;)

> 2. In the event that I get accepted as a student (for something else, but
> still KDE) would anyone be prepared to take over mentorship for Knights? It
> uses the KDE technologies as much as possible, so there shouldn't be any
> unknown code in it. The code base isn't large, and I tried to document its
> design, but if there are any questions about the code I'll try my best to
> answer.

The first part of the task (saving games to file) is quite straight
forward and could easily be mentored by anyone with basic Qt
development experience. The second (and in my estimation, much larger)
part, however, I think requires quite a bit of high-level chess
knowledge. As a non-chess-player, I really don't know why one would
want to analyse a set of positions and how one would best do so. I
think this would severely limit my ability to mentor the development
of such a UI, so I think you'll need a mentor with more chess
experience.

Parker