That's probably true, but it depends a lot on what features are needed.

Because i'd be happy with a clean, attractive display of the text with a very simple interface to highlight portions and add annotations, the browser seems like it might be the shortest route to a very polished product. If i was trying to compare texts side-by-side and do a lot of dictionary lookup, i'd definitely not waste effort with a browser.

The thing is, i get hung up on the portability of the annotations. I don't want to lose them on my desktop or in a few years as tech changes, and that's the real challenge. A layer that sits on top of a standardized browser platform seems like it has a decent chance of being portable to desktop, and future tech...

Incidentally, that's why i asked ARJ about whether he thought the ideal interface for a maemo device used just a single display (rather than split screen): if the IT is too small to effectively use a split screen, anyway, it makes the browser approach more appealing.

Edit: the above was written before the last post. It sounds like you're confirming that the IT is just too small to justify a split screen UI?