H.G.Muller wrote:I am not sure what your objection is. If you click a PGN file it automatically starts WinBoard with an extra 'game list' window, which stays open all the time, and where you can see a directory listing of all the games in the file. Just click on a game and it will load that game. Games are allowed to begin with a setup position, so you could put games of zero moves starting with the problem position in there. This all seems very convenient and user friendly. Your students could do the problems in any order they want, and switch back to those they already did before.
If you are going to change the code, keep in mind that I am just about to relase a version that is so heavily changed, that any derivatives you make from older sources will be dead branches from the start.
I'm teaching 5 and 10 year olds and I am teaching between ten to thirty at a time.
I would really like to be able to drop them in front of the computer and not have to explain much or have to worry about them for ten or fifteen minutes. The more they can do on the computer without my help the more time I can spend with other kids.
Your idea could work, but I just see myself having alot to explain and alot of supervision. Maybe they will need to know why their move doesn't work, or they accidentally closed the game list and need me to reopen it for them. But mostly I see them staring at the screen having no idea what to do, even though I explained it to them twice.
If instead I could load up a testing engine (that I wrote), then the whole process could be alot more interactive. It loads up a position and they make a move. The engine shows why the mate isn't mate by escaping mate, waits five seconds and resets the puzzle. After several wrong answers it skips to the next puzzle. If they are taking too long I could even pop up hints to help them along. This way will also require me to watch them a little, but probably alot less. Also they will be able to go through alot more puzzles alot quicker.
I also think it would be alot more effective if the engine could randomly choose them rather then the student trying to remember where he left off last week. Also for the positions with multiple variations, this could handle moves not expected. Also maybe we want to make sure they are prepared to handle all of the variations. If it is a mate in three, after they get it right, take back a few moves and make sure they can mate if the opponent chooses another move instead.
I could see this ability being very useful to alot of people.
H.G.Muller wrote:If you are going to change the code, keep in mind that I am just about to relase a version that is so heavily changed, that any derivatives you make from older sources will be dead branches from the start.
Winboard seems to work pretty well already. What new features would I be missing?
Is this feature that I am talking about very difficult to implement? I am imagining that it would not take more then a few minutes to add (for someone very familiar with the code). Would you mind putting this in the next version?
I may end up doing this myself, but it would take me a considerable amount more time to do this. I would spend at least a week, probably more, just trying to get familiar enough with your code to know what to change. I would also spend a week just trying to compile it. For some reason my compiler refuses to compile code written in a earlier version of c++. I need to get the exact version or it gives me alot of errors. If you are making a new version anyway maybe you could just toss it in?
Think about the usefulness of this feature. How many people are writing programs that need a chess board besides chess engines. Alot. And if this feature will allow them to use Winboard instead of writing their own you will save that person alot of time.