Han-Wen Nienhuys <hanw...@gmail.com> writes: > There are two separate discussions here: > > - how do we offer to average user a way to extend the program. I agree > that C++ is not the way to go > > - how do we offer developers an environment to extend LilyPond, were > extensions go back into mainline; this is connected with getting more > developers on LilyPond.
You can't separate the two. Developers grow from users. Look at the TeX/LaTeX and Emacs communities: how much of the changes happen in the binary, how much in the interpretative layers? Where did most developers get their first experiences and contact? >> Scheme might not have been the optimal choice, but it beats not >> having an interpretative layer, and our interpretative layer is still >> much more limited than desirable. > > To me the question is where we should invest: having the interpretive > layer be more rich (where it is already incredibly rich *and* > incredibly obtuse), How about making it less obtuse? > or having better fundamentals (page breaking, spacing, collisions > etc..) That's a red herring. To work with fundamentals, you need to be able to juggle and express them in a convenient way. Juggling is easier done in an interpretative layer, and the expressiveness is a matter of designing good programming interfaces. Scheme is excellently suited for tying together suitable functional building blocks for linear programming, arguably quite better than C++. But one needs to prepare the fundamentals in a way where Scheme is a good fit, and not tack on a Scheme layer as an afterthought. -- David Kastrup _______________________________________________ lilypond-user mailing list lilypond-user@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-user