Dear list,
I get jealous when I see SuperCollider users sending around one-liners
like:
play{SinOsc.ar(OnePole.ar(Mix( LFSaw.ar([1,0.99],[0,0.6], 2000,2000).trunc([400,600])*[1,-1] ),0.98)).dup*0.1}
Having a scripting language front-end is a great way to share snippets
of code on social networks, twitter, chat and the like. To my
knowledge Pd has nothing like this; the only way to express a patch is
graphically, right? This got me thinking about alternative (textual)
ways to express a dataflow/DSP graph.
Firstly I'd love to know if anyone else has worked on this. I'd also
be interested to know if there are any existing *textual* dataflow
languages that could be useful here.
Anyhow, I've had a first stab at a textual representation of a Pd
graph. It can be found at:
http://www.jamiebullock.com/2009/04/pdscript.html
Thoughts welcome!
Jamie
Jamie Bullock wrote:
Firstly I'd love to know if anyone else has worked on this. I'd also be interested to know if there are any existing *textual* dataflow languages that could be useful here.
I played around a bit with Haskell's Arrow notation ( http://www.haskell.org/arrows/syntax.html ) a while ago:
fmExample = proc _ -> do
modA <- osc 0 -< 110/8
modB <- osc 0 -< (modA * 110) + 220
carrier <- osc 0 -< (modB * 220) + 440
returnA -< carrier
Also Faust ( http://faust.grame.fr ) seems interesing, although most of the documentation seems to be PDF.
Twitter copy and paste app for Max/MSP;
http://www.soundplusdesign.com/?p=1621
Somebody port to Pd, or do something similar?
On 20 Apr 2009, at 14:07, J. Simon van der Walt wrote:
Twitter copy and paste app for Max/MSP;
http://www.soundplusdesign.com/?p=1621
Somebody port to Pd, or do something similar?
I'm sure something similar could be built with Pd. My problem with tas
is that the notation is way too low level and lacking in semantic
information.
Porting this to another platform other than Max in a meaningful way
would IMO entail copying the implementation. I would prefer to see
something that defines a protocol, API or scripting language where the
data representation is decoupled from the implementation.
Still, I think tas is an interesting project, and it *works*, which
is good in itself.
best,
Jamie