[CM] Building applications with CM?

Rick Taube taube@uiuc.edu
Sat, 9 Sep 2006 08:58:18 -0500

> What possibillities are there to develop "Applications" based on  
> Common Music?

if you look on cm's sourceforge File Release page you will see that  
for osx i save an "application" version of the stable cm branch that  
works out-of-the-box if you have Aquamacs Emacs installed. (I dont do  
this for linux because these folks are generally comfortable  
assembling software  from scratch). this could be done for the  
unstable-branch as well if some mechanism were added to auto-download  
newer sources under the application (M-x update-system cm clm ) ??

> The Idea is to build something (with some kind of gui maybe) i  
> could pass on to my friends to play around with,
> without having them to mess with stuff like Lisp compilers, the  
> common music sources etcetera...

A laudable goal, however Emacs and Lisp syntax will be HUGE hurdles  
for people (non-technical composers) that want to learn about  
algorithmic composition, in my opinon your wish to "pass on to my  
friends to play around with" is very optimistic... I have very   
intelligent students that -- even after a semeseter's study -- are  
still frustrated trying to express even simple ideas in the emacs/ 
lisp environment. but go for it! the holy-grail would be a cross- 
platform, non-emacs "gui" that integrated graphics (plotting and  
notation) with text-based algorithmic development with/without lisp  
syntax. A few years ago I had really hoped that gtk would be the  
answer for the GUI side and i actually started this work, but even  
now there is still no native version on osx and i wont do any more  
gtk until i see at least that (in my lab cm's gtk gui takes more than  
a minute to open a window, its a joke) SO im in a holding pattern for  
gui work, as usual. in the meantime two true cross-platform gui tool  
kits have surfacec that might allow some way to move foward: wx and  
JUCE. JUCE look particularly interesting. and FOMUS does an amazingly  
good job of translating generated events into a notation. so perhaps  
the pieces are there.