I was just readin some lists about sound, MIDI, and OS X and I read this:
"According to various beta-testing mailing-list of major editors
(Emagic, Steinberg, Native-instrument,..), programmers all say the same
Yes OSX seems great, but we won't develop something before Apple
communicates us the standards about MIDI and Audio programming. At this
time, there are no guidelines or API, and we won't waste our time and
money to re-invent the wheel. So we are waiting. "
Maybe this is our opportunity to get the MusicKit adopted by some major
developers. I think we have a chance of getting some of these companies
to use the musickit. Here are some selling points of the MusicKit:
its API is stable and established. Since we are version 5, the
developers wouldn't have to worry about developin using a API that is
still under dev.
Its OS X/OPENSTEP/Cocoa native. The MusicKit was developed on NeXTSTEP
until v. 4.2. OS X is really just NEXTSTEP 5.0, especially when you
consider that NeXTSTEP 0.x could run Mac apps.
Native DSP support (soon!)
Simple port to GNUstep (soon!), therefore linux, etc.
Anyone good at sweet talking CTOs?
"Man has survived hitherto because he was to ignorant
to know how to realize his wishes. Now that he can realize
them, he must either change them or perish."
-William Carlos Williams
From: Leigh Smith <leigh@to...> - 2001-06-07 16:59:19
Midiman are now showing some interest in the MK and thinking of bundling it
on their driver CD.
We actually now have Linux looking pretty good, although there are a couple
of GnuStep compiler issues that need fixing and a working MKPerformSndMIDI
before it really flies.
I've just finished a bridge to Squeak Smalltalk, so you can play Scorefiles
within Smalltalk (think of it as interpreted ObjC). There is still an issue
with separate threads that doesn't manifest badly in 5.3.0, but has held up
a 5.3.1 release as it really needs a factoring in MKConductor.
Frankly DSP is still quite some way to go, but this would make an excellent
student Honours project and make them eminently employable.
My general feeling of other developers are they are now seeing the problems
of their designs in attempting to port, having been in this very privileged
position of writing nearly to raw hardware in the case of Win < 2000 and
Mac < 10. Clearly one+half programmers (Skot and myself) were able to
achieve a functional system just by reading header files, without having
ADC premier access or special hand-holding.
I'm going to be away for the next couple of weeks, so things will probably
be a bit quiet. Anyone is welcome to pick up the slack :-)