I think that you're in danger of spreading yourself too thin and would counsel doing one thing at a time and doing it well. We know from experience that a big
You asked ;-)
I am glad you said this. My whole work around lyx is making lyx more user-friendly. At least presently, I am not touching (and I do not have the needed knowledge to touch) the core of lyx. My work is not even guided by the bug list, rather by my own observation, experience, and mostly frustrations. I have already several patches. They are almost done, and I can promise you that I will take my responsibilities to finish them. I am not working on them right now, partly because I would like to clear them one by one. (BTW, can win/auto-view go in?) Now, the biggest problem of lyx, to my understanding, is the portability. Building lyx on windows or cygwin has been a pain (my own experience), and this will certainly deter potential contributors and users. As lyx grows bigger (e.g. +qt4) and needs better support on more platforms (win/cygwin/*nix/mac), auto* is (obviously to me) a deterring force, not a helping one. That is why I am looking at scons, which has helped other open source projects solve the portability issues. Another private reason is that I am also looking for an auto* alternative, since I have been tortured by auto* long enough for some of my own projects. Now, regarding the possibility of replacing auto* with scons, I agree that there will be a lot to do to implement every single feature of auto*. I, however, will soon enjoy the speedier building process under windows. This, to me, is a well spent of time. (-time, +learn something better than auto*, +future saving of compiling time.) If other developers would like to try, I am more than happy to send them the patch. If people like it, we can put it to trunk; then, we will have a chance to implement more features and maybe replace auto*. Cheers, Bo
