Re: GSoC idea: Gnome Games + Telepathy Tube

[Zhang Sen <zh jesse gmail com>]

Hi Thomas,

On Wed, 2009-03-25 at 16:19 +0100, Thomas Dybdahl Ahle wrote:
> > 3. Searching 'collaborative sudoku' on Google gives several other
> > interesting results, e.g. a paper [3] inferring peer-related emotions
> > (using a collaborative sudoku), and also a workshop [4] relating to
> > collaborative sudoku.
> 
> Interesting. How did the paper view sudoku as a collaborative game?

In the page, a pair of couples compete with each other. The two students
in one couple will collaborate. The couple work on a shared grid, and a
numberbox that is filled can only be changed though negotiation. In this
process they investigate peer-emotion.

> For competitive mode we also need to find out: Should players play the
> same sudoku? And if they do, how much information can be sent so that
> the other player can't cheat? Otherwise gnome sudoku has functionality
> for rating board difficulty..

They should play the same puzzle, since one player will be initiator,
whose sudoku decides which puzzle to play and send it to the other
player. As Thomas Hinkle said, this can be made a multi-stage task, e.g.
first be able to share a puzzle (the friend trying to beat the time),
then to lively update the other's status.

> If one grid could be made fluent, it might be quite fun. Perhaps not for
> serious solving, but at least to try it out.
> How do you imagine two grid collaborative mode working? What is the
> collaborative part? That you can view what the other player has solved?
> Perhaps then your own grid can have small hints on fields for witch your
> team mate has a certain solution.

Using two grids, collaborative mode will be the same to competitive,
except your teammate's numbers are shown.

Using two grids both for completion and collaboration is more unified
and needs less work. Collaborating in one grid seems more interesting,
but also a little radical.

So, please give me some suggestions as to,
1) is one-grid or two-grid simply preferred over the other, by
experienced sudoku players and experienced developers? If so, the
solution is straightforward.
2) does GNOME require/want the unification and simplicity of two-grids?
(If so I will settle on the former)
3) can the workload be estimated? If one-grid collaboration is not
likely to be done well within a summer (based on past average GSoC
students' ability), then the former.

Or maybe I asked the wrong question? :) Any comments welcomed.

Cheers,
Zhang Sen