Re: [Strasheela-users] strasheela

[Torsten A. <tor...@pl...>]

Dear Olle,

I feel from your previous post I got some idea what you are after,  
and it makes sense to me. I see Strasheela indeed as an algorithmic  
composition or computer-aided composition tool, but one which is  
particularly good at a particular technique of computer-aided  
composition. Common Music is very good at (nested) patterns, Symbolic  
Composer has lots of ways to re-interpret the meaning of its symbols  
(a, b, ...), and Strasheela is good at using constraint programming  
for music composition (i.e. defining interdependencies between  
variables). Still, as the Lindenmayer-system example on the  
Strasheela site shows, I don't feel that constraint programming  
should be the only way :) On the other hand, I feel I must not try to  
compete with the other systems: I can not incorporate every technique  
(at least not myself :) ). For example, Strasheela provides a rich  
set of patterns (see the Pattern functor). One the one hand, many  
patterns are more expressive that CM's patterns, because they are  
constraints. On the other hand, the Strasheela patterns can not be  
nested yet (unlike CM's pattern): I simply did not find out yet how  
to do that in a constrainable way.

I fully understand that Strasheela is a system which is somewhat hard  
to get into. During the design, my primary goal was to make the  
expert-user happy (much like systems as SuperCollider do), but nobody  
starts as an expert user :)  We are working on this matter by  
improving the documentation etc...

> Thank you for letting me take so much of your time.

Thank you for your interest in our work :)

Best
Torsten


On Jan 31, 2008, at 11:55 AM, Olle Romo wrote:
> Hi Torsten,
>
> Thinking it over a bit more I suppose what I imagine is to use  
> Strasheela in an algorithmic composition way but with constraints.  
> I'm thinkin' I can specify intricate and interesting  relationships  
> between notes and parameters and arrive at parts where aspects tie  
> together in a way I find almost impossible to achieve with other  
> tools.
>
> I'm not a fan of Common Music's, Symbolic Composer's, AthenaCL's  
> and others use of randomness to generate variations. I'm intrigued  
> by (the thought of) the ability to specify interdependencies and to  
> generate solutions to that.
>
> I understand now that algorithmic composition is not the prime  
> reason for Strasheela and I think I've been searching for something  
> that's maybe not implemented yet. I've been thru your thesis twice  
> and spent a fair bit of time with the toots and examples but I'm  
> still stumped as to how to achieve what I want :) I'm deeply  
> intrigued by the system and I see some very exciting possibilities  
> if I just can get a handle on some pop related techniques.
>
> Maybe you know someone who would be willing to tutor me? I will  
> happily compensate for any time spent. One day I will be able to  
> contribute to Strasheela, but until then... :)
>
> Thank you for letting me take so much of your time.
>
> All the best,
> Olle


--
Torsten Anders
Interdisciplinary Centre for Computer Music Research
University of Plymouth
Office: +44-1752-233667
Private: +44-1752-558917
http://strasheela.sourceforge.net
http://www.torsten-anders.de