Notices

So after another long nights work, mikhas and I managed to pack a release for you guys to try (I think it is not released yet, but should be tomorrow). As you can see in the screenshot, all pieces are up and running with their move rules implemented. Also black and white take turns and pawns are promoted to queens.

Things to be done are: check, checkmate, all sorts of draws, castling and en-passent.

Not yet, according to http://maemo.org/packages/view/miniature/

/me is impatient! Thank you for the hard work.

I can donate some iTunes credit if any of the guys working on this would like to download a bunch of chess apps for the iTouch/iPhone for testing.

I think T chess (forgot the full name) looked pretty good.

Anyway, it might help to improve creative methods on how to best have a friendly UI on a limited display screen.

Addison, I don't have an iPhone/iTouch but I would really appreciate your feedback on best implementations out there.

Also, what about having releases for Maemo 5/N900 and Ubuntu at the same time? If it's not much extra work. It is useful to get testers (now most people in this thread don't have an N900) and also good to keep in mind the cross-platform strategy, even if currently having a full supported desktop version is not a goal.

To me, there are three important things to understand before taking on any type of new project. And that is research, research and research.

That all night, coding marathon on programming legal piece movement has been done before, several dozen times over by others, and I'm sure that the open source code for this was already available somewhere just looking to be found.

I think someone should step up and volunteer as a sole researcher for Miniature. Find parts of open source coding that is currently needed, play virtually every available chess program that could potentially fit within the framework of this project, and document all records if certain coding is being used from another author. You know, junk like that.

Maybe I missed the post somewhere but I don't remember reading any type of research that has been done already. It will save time in the long run.

Don't look at me to do this though.

I have no intention of getting a N900 and I wouldn't have the first clue how to run Ubuntu on my computer.

I can donate some iTunes cards though if needed. It's my wife's iPhone that I don't ever use. If no one else has one either I guess I could always ask her if it's okay to borrow it for awhile.

By the way, I think that this was the program that Miniature could borrow the most from since it features finger friendly touch screen controls and a dynamic UI. No stylus needed.

T Chess Pro

You are probably right, the stuff we've been coding the last few days has pobably been done loads of times before. But, honestly, it would spoil most of the fun to just look around for reusable open source code and copy&paste the hell out of it. Of course it would probably be faster but I really enjoyed what we've been doing so far and I don't think I would have joined this project if we would just walk around like bums searching trash cans for things others might have done.

Ah if software development was only that easy ... take that component from here, then that one from over there, then wire it all together in some designer tool, push a "make it so" button, release.

There are many reasons why it doesnt work like that, but I'll cut that out. It doesnt belong here.

Also, your link points to a commercial app (admittedly, it looks nice!), so how am I supposed to reuse code from that?

I am not even touching the issue of code portability, given that the iPhone lives in its own Objective-C world whereas Maemo's future lies in Qt.

You can be assured that I *did* look at code from other FOSS projects, and while doing so I changed my mind several times but we eventually decided to focus on writing our own move validation engine, simply because we want a component that integrates nicely with the rest of our program. Think of the UX, even a simple chess board needs to know a lot about chess if it ever wants to present useful feedback to the user. For example, take the dots in that app's screenshot, showing the possible moves. How does the board know where to draw them?

Nevertheless, we still dont aim at writing our own complete chess engine, that'd be insane. Again, look at our roadmap!

History shows that this is true and not. There is also plenty of very good research that ended up nowhere. There are also many examples of deep research that knew so well the precedents that could never escape from them.

I think we have a good proportion of research, user feedback and development. These Talk threads are being quite useful keeping this balance.

Welcome!

Doh, I got too enthusiastic too soon.

This program is for "Play against computer", which is our item number 9 in the roadmap.

http://www.iphone-game.com/images/st...%20pro%201.jpg shows an implementation for selecting a piece, which is something we are asking to the designers around. Apparently mikhas & kore want to try the idea of selecting a destination square first and then highlight the piece(s) that could reach it, as a way to minimize false moves.

They have decided to make the board smaller, adding the coordinates outside the board. Maybe good enough for analyzing games but for playing any pixel counts.

If someone could dissect the settings menus that could be useful.

The UI for playing games is more austere. Traditional notation (what do you think about my proposal for using natural language?) and of course no chatting since this is quite pointless with a computer (unless you reach HAL levels of AI).

The user comments seem to be very good. If that is the best chess game for the iPhone then I'm confident the Miniature plans are on the right path. When someone plugs the Scid backend to the Miniature UI we will have a good contender.

A new .deb release is available: http://maemo.org/packages/view/miniature/

There is a simple route to enable online chess for the n900, that is to enable Java for the n900. This would have many spin offs, my last mobile a Samsung e900 seemed to have no problem with Java, but you could not install apps like Jin because of memory requirements. Enabling Java for the 900 connects both ICC and FICS to the 900 in terms of chess on the 'net and allows a myriad of other web apps to be accessible by the 900 also.

All we need to look into is, acquiring the source code for Java, find a competent coder and I possibly know one, compile a binary, ride the wave and Nokia has a winner BIG time.

Regards,

Peter.

http://dollyknot.com