a better strategy would be to target particular users who only need one application that is almost working. At least, that's what the model I wrote told me: http://yokozar.org/blog/archives/48
That strategy(to no ones surprise) appeals to me, since it feels like common sense, and is close to how most of those I work/worked with think. But I am not sure it would have worked earlier in the project when having nearly working use cases would be less usual and when having broad, or no focus would make the project more interesting for its developers.
Anyway, It seems likely to make the most number of users happy with the least amount of work. However, since some applications(no names) are very widespread/pirated AND close to working, it will have the effect of more or less officially focusing on certain applications, which i think would be hard to push through in this project for different reasons. To try and define the most usual "near working" use cases also means defining the most usual use cases since it then would be important to keep the "working" use cases from becoming "near working".
To do that, one need needs user input which would make the project user centric. Of course, the *entire* project wouldn't have to have the same focus but it would be affected. I'd think that this would be good and appeal more to me and others as developers wanting to do good things(tm). But I know that many does not agree.
//Nicklas