NET-MIDI idea

["ROSSI J.A." <rossi@nusc>]

About 3 months ago, I attempted to post this idea to Love-Hounds, but it
got lost in the net-ether.  The recent posting ny Peter Alfke concerning
synthesizers and net people prompted me to again try and post this.  After
going through the 57 mail messages I had waiting for me this morning, I've
decided to give it a try again.

Expanding on Greg's Net-tape idea, I thought it would be kind of neat to put
together a NET-composition.  Now, in pre-MIDI days, this would be totally
outa-touch, but given that this IS a computer network, and MIDI data is,
in fact, data, I feel that a composition en masse is not only possible, but
also doable.  Here is an idea of one way it could work (I would also
appreciate any input as for improving suggestions)...

First, several things would necessarily need to be worked out before the
actual realization of the composition.  Basic things such as rhythmic
structure, meter, general description, etc.  It would be interesting to
work out some mailable notation system, but I'm not entirely sure that it
would be necessary.  As soon as the initial barriers were passed, it would
be possible for individuals to work out actual voices and sequencesl.

As each part is worked out, it could be sent out over the net as a MIDI
data stream.  Since MIDI data is restricted, for the most part, to 7 bit
words, it should lend itself well to ascii coding and decoding techniques.
Because we all probably have access to differing brands of MIDI equipment,
passing system exclusive information is probably not a major consideration.
Now, I realize that a lot of stuff is going to have to be passed as verbal
script, but I feel that the subjectivity which would be added by such a
scheme would add to the realization of a non mechanical piece.  For example,
a person with a MIRAGE could send data which would be playable on a CZ.
Although the voices would be obviously different, if the MIRAGE player stated
that the voice 'sounded like' (or in the case of a Mirage 'was') a 14ft grand
with reverb, the CZ player would set up as best a reverbed piano as possible
to listen to the basic idea.  The CZ player could then consider layering
another voice on the first.  As the generations pass, the polytimberal
structure should eventually evolve.  Decisions concerning the actual
composition could be made at various places along the way by actual listening
to the 'current composition'.

At any rate, I would appreciate suggestions concerning this idea.  Do you
think it is off the wall?  Is there any enthusiasm out there for this sort
of experiment?  Send me your comments, etc.

Finally, Peter Alfke mentioned rec.music.synth.  Is this a mailing list,
user group or what?  How do I subscribe, if possible?  Any information would
be appreciated.

Comments to ROSSI@NUSC.ARPA .....

John 
------