Actually, assuming we can / do create a single application for iPhone / iPad, we wouldn't even be able to charge you for it twice; if you attempt to purchase an add-on that you've already purchased before with the same iTunes account, iTunes detects that and offers to let you activate it again for free.
(but to anyone reading this, we'd really appreciate it if you'd restore old purchases using the "Restore iTunes Purchases" function in Settings / Registration instead of by buying them again - it works just as well, but guarantees that you won't accidentally buy something new (say, the non-educational version of a product you previously purchased the educational version of - Apple treats those as two separate items), and creates a lot fewer accounting headaches for us)
Test administration would be an interesting application for Pleco flashcards, though at the moment it's far too easy to cheat - we could design a UI that didn't allow that, though. Multi-choice would be too easy / inconsistent, and self-scored is obviously out, but a free-answer test (characters / Pinyin / tones) could be interesting; maybe also include an English option (even if we don't determine ourselves whether it was correct / incorrect) and record the stroke order they use when handwriting characters, then provide instructors with a way to retrieve / review / revise scoring for the results.
I've long thought stroke order testing specifically would be a good application for this - grading it ourselves would only work with the current tap-in-the-outline UI, checking / matching strokes as they'd drawn is just too unreliable and has too much potential for false positives or negatives (which are a big no-no when people's grades are on the line), but even just recording the strokes they drew in the order they drew them and letting the instructor review them would offer something that's completely impossible on a paper test.
How much time WM7 takes up is still very much up in the air (as is the question of whether we'll be able to develop for it at all, I suppose), but given that it could be another 6 months before we see it shipping on devices we don't necessarily need to get started on it the second they release a developer kit.