Originally Posted by ProDigit
However it would make it hard for the device to read that input, so the magnetic input would be limited on the screen, and can not be saved into the memory.
Well, it would depend on how it was implemented.
If the display were built such that the processor could go back and "read" how each pixel was "set" (black or white), then it might be able to read it back out as a bitmap. At that point it might be possible to, for example, filter out the original "image" from the bitmap and store the remainder as notations that could then be overlaid back on the page when it was displayed again next time. That would most likely require a more complex display structure, but then again, it would also require a totally different
structure from what they're using now.
As for future development ... I like the idea, obviously, since I've spent a fair bit of time kicking it around, but I'm nowhere near a place to be able to answer whether it would be feasible or not.