I've been thinking about Go software currently available for smart phones & a/i-pads. I had what I think was a good idea, but there's no hope I'll ever have the time to implement it (and I only program in FORTRAN & IDL in my job). - so I thought I'd post it here for discussion / interest.

Basically the idea/scenario was:

"I open up the book I'm currently studying. Page 3 has a diagram with 50 moves on (for example). I use my tablet pc/phone to take a photo. Then on screen I crop to the diagram and the software converts it to sgf. (I know this has been done before, though not on mobile devices). I can now loop through the diagram and watch the sequence unfold.. OR I can learn the sequence using a method like Daily Joseki uses (great website)*."

* I find it hard to pay attention to sequences until I've played through them repeatedly. Plus you could then test yourself later with these sequences.

so you get diagrams like you would have in smartgo, but without having to pay twice for books and with no limitations on the books available (plus I had in mind it would be on Android OS which is what I have). - I'm sure some people would prefer to use smartgo books though being more polished and not making the reader shift from looking at a book and then a screen etc.

Secondly, I thought the touchscreen capabilities of a tablet pc could speed up creating sgf problem files, perhaps an application designed to make it really easy to do this (using the image capture-> sgf function of above? or just painting stones on with fingers).

I suppose it's possible for someone to say you should just play the moves out on your physical Go board, but digital learning offers some advantages IMO, like the possibility of spaced repetition. For example, I've been studying Joseki (for fun) for the last couple of months. For me, the best approach combines Daily Joseki study and book study (neither would work as well for me).

Of course if anyone has the time and inclination to create such a program I would be very happy to pay for a copy - I doubt it would make anyone millions of pounds but for a serious Go book nerd it could be quite handy.

_________________
I am John. John-I-Am.

Could you elaborate on this, please? It sounds like scanning a go diagram. Past experience with this seems to have produced abysmal results even when it claimed to have worked (e.g. columns of stones shifted one or two lines), but usually it hasn't worked anyway. Apart from different fonts, fuzzy numbers and other typographical variables, there have been problems with captured stones, kos, throw-ins and so on. As regards doing a complete game record when it is split up into several sub-diagrams (some of which may even be abbreviated), I'm not aware that anyone has attempted this yet.

Or have all these problems been solved?

On a related aspect, Bill Cobb of Slate & Shell made available a free download of an sgf file of all the variations for the game in one of the books I produced (I forget which one). It was mentioned prominently in the book. He told me at Santa Barbara that next to no-one has availed himself of this file. Both of us were surprised, and he naturally seems disinclined to continue the experiment for future books. Possibe interpretations are that people who buy these books are able to manage with the profuse diagrams Bill provides or that people who buy books are not inclined to play over the games on a computer, or are there better explanations?

I would fall into "people who buy books are not inclined to play over the games on a computer" group. It's easy enough to just follow the book diagrams.

ah sorry - I think you probably know more about it than me. I don't have any practical experience in this but I was thinking of a few similar sorts of projects I've read about over the years. So those problems have probably not been solved. - though I thought Anders might be doing something like that in producing the Smartgo books.

However, I don't see why it would be difficult to mesh two sections of a game in separate sgf files together since I would have thought it would simply be a case of finding the point where the board configurations are identical (assuming no errors ).

_________________
I am John. John-I-Am.