On 07/09/2012 03:17 AM, Element Green wrote:
Having said
that though, I don't personally see an issue with implementing OSC SYN
support in FluidSynth, provided there are actually applications which
could interface with it in a usable fashion. It could become a
compile time option, for those who don't want to build that
functionality.
If someone came up with a working and clean implementation of an OSC
SYN enabled FluidSynth, I doubt anyone would object to merging the
support into the main project.
I don't know much about OSC or OSC/SYN, and have not used it myself. And
that's part of the problem, really, because I don't know what it can do
for us, how common it is, how standardised it is, and so on.
But we are a library, and as such, for us to be successful, we have to
be useful for other people and applications - the more the better.
OTOH, we don't want to merge something, see the original contributor run
off, and after a few years with some refactorings inside and so on,
maybe the OSC support breaks and we're left with bug reports for stuff
people don't know how to fix (because none of the original developers
use OSC), and FluidSynth losing reputation as a result.
Not unlike suggested by others, I think we could start as something on
top, i e, as a binding layer that calls into the fluid_synth_* methods.
The OSC layer would be calling these methods, and we would be open to
take patches that makes support for OSC/SYN easier/possible, if that is
necessary. As both Element and Pedro has pointed out, FluidSynth API can
do more than what is possible to express with MIDI already, but there
might be gaps which we discover now - or along the way.
If it then proves to be reasonably mature and bug free, useful and used
by more than one person, and we have people on this mailing list that
can support user's questions and problems in the area, I think we should
merge that code.
Does that sound reasonable?
// David
_______________________________________________
fluid-dev mailing list
fluid-dev@nongnu.org
https://lists.nongnu.org/mailman/listinfo/fluid-dev
