[PD] setable values

[Tim Blechmann]

> As I understand Miller's motivation, this option would only be for
> things like developement. THis way you could test some changes to
> builtin objects without having to recompile Pd all the time. 
recompiling pd is something that can be done within one or two minutes
... when you are working on an external and only changing one file, you
just have to recompile the file you changed and link the objects ...
thanks why there is the make tool ...

> Currently we have another problem: That there are 2.5 parallel
> versions, which are neither older or newer than each other. (Or you
> could say, that currently Miller's version is the newest, this might
> change again this weekend. ;)
since i commited miller's changes to the devel_0_37 branch, it is the
newest one again ;-)

but you are right ... having two or three branches is not very good for
the development... it's not really good if there is one version with
development in the dsp core (msp), one with development in the gui
(impd) and one with different additions (devel_0_37).
basically miller is working on the core, mathieu on the gui, and a few
other people like thomas, guenter and me on some other features ...
we all agree that miller's branch is the official branch, the language
definition. 
i don't know, if it's useful to have mathieu's and miller's development
divided to two different cores... if mathieu is working on one part of
the main branch (the gui) and miller on the other part (core), they
should work on the same branch.
also things like thomas's inlined simd code (which is far better than
anything gcc or icc produced, i checked the machine instructions and
did some profiling), like guenther's tooltips (which are _very_ useful
for to use in abstractions) or my fftw support (more than twice as fast
as the mayer-fft, and selectable at compile-time) or threaded soundfiler
(which has been reported to work on win, osx and linux) or yves patch
for a threaded gui (i have to use it for my performance patch to be
click-free)...

and i doubt that we should reinvent the wheel several times by rewriting
something that is available in another branch...

cheers ... tim, who is hoping to see one stable branch and one
development branch that will be a future stable branch...

-- 
mailto:TimBlechmann at gmx.de    ICQ: 96771783

After one look at this planet any visitor from outer space 
would say "I want to see the manager."
				      William S. Burroughs