Miniature has some source code already!

http://gitorious.org/miniature

I guess only Mikhas knows whether it is functional at all, though. If you give some instructions about what to do in order to get something running on the N900 I will follow and help with testing and feedback since the earliest stages. I'm not familiar with git repositories and etc but I'm willing to learn.

About the arrows. ok ok, let's give us some time and tests to see what are the right defaults and alternatives (if any) for representing moves. Showing legal moves is not even in the roadmap yet.

We just need to find the right balance between all the things that could be done, the priorities to make happy the first wave of users, the resources we have, the simple/complex UI we wish for settings etc...

joshua.maverick, thanks for the offer! There is two things you can start with:

- We need a logo/icon. The application is called Miniature, I suggested we could go for something based on a humble and simple black pawn. A starting point could be http://gitorious.org/miniature/minia...black/pawn.svg - oh sorry, I mean http://upload.wikimedia.org/wikipedi...hess_pdd45.svg . But feel free to go your way as long as you keep a clean track with copyrights / licensing.

- You can help creating a nice-looking yet simple and very usable UI for Miniature. See http://wiki.maemo.org/Miniature#Mockups - even if the sketches there look nice, they are actually not intended to represent final look&feel. The attention now is put on the UI layout (what elements are where) and the usability.

I recommend you not to touch the board/pieces themselves by now since we are going first for the very standards. Custom boards and pieces appear later in the roadmap and even then there is a lot of freely available artwork.

Let us know if you have other graphical ideas.

On it. I'll spend some time catching up on this thread, and start working on designing stuff as early as wednesday. Sounds like this is going to be fun, I'd work now but 5 hours sleep in 72 hours is a bit much

So actually, we could really need two polished boards (one portrait, one landscape), with a *fixed* pixel size

Having the cross-platform compliance in mind, wouldn't it be better to avoid completely fixed sizes, offering always a fluid UI? Alright, something like an avatar might be an exception but the board willing to take as much real state as possible on different form factors might be a problem.

What if we discover at some point that we just need a recompile to have Miniature running in a Symbian, Windows Mobile or Moblin device that has less/more than 480px screen?

Also just checking: are we doing things right to allow localization when the right time comes?

Also just checking: are we doing things right to allow localization when the right time comes?

Yes, you are absolutely right. I didn't think about this properly. So the question is: Is it possible to create a crisp "as-if-it-were-fixed-size" looking board while using SVG?

I am porting my simple PGN viewer to Maemo, and it's almost finished. It does nothing but display PGN files, but it looks good and has really good finger usability. I will blog about it as soon as I find time to finish it up, but these shots should already give the idea:





Those are SVG pieces, so they can be used for any display configuration. Feel free to use them. The board is scalable too, but I'm using pixmaps for the wood grain pattern.

It's an interesting idea to have a chess client just for displaying games. Perhaps it would even be a suitable addition for Chessmonk (which is the name of my utility), I will have to think that over. I also had the OCR idea right before I saw qgil mentioning it. I wonder how reliable it could be, but it's definitely worth investigating. It would have saved me a lot of hassle when I was still studying the game.

I believe that making a good app for chess playing on the phone will be much harder, partly because the complexity of chess servers can't simply be ignored (you may get in trouble with other players if you do...). But also for the reason you mentioned, that it requires really fast and precise input. Good luck to anyone who tackles this problem though.