On Tue, 2006-09-05 at 12:51 +0200, newsletter_at_t-klein.com wrote:
> there's a difference between 'natural' and algorithmic thinking you
> haven't considered so far: feedback
>
> When you play a guitar, you get direct physical and instantaneous
> feedback from your fingers, which are also your input organ.
> On the other hand, programming can only give you non-instantaneous
> feedback (whatever small the delay might be), which doesn't involve your
> input organ (fingers), but rather other senses (visual, acoustic).
I think Jeff Pressing's model[1] of how people improvise comes in useful
here. He characterises an improvisation as a series of clusters of
events (clusters of sounds). A cluster ends and a new one begins when
the improviser decides to make a change to how they are playing -- he
calls this moment the "interrupt." The actual generation of events
follows from this decision. The decision is expressed as modification
of arrays of what he calls Objects, Features and Processes with some
reference to ecology to suggest how these arrays may be built during
one's life. He also describes how feedback from the real results of
physical actions enacted by these events later influences following
decisions during an improvisation.
So in this model a livecoding improviser is not involved in the physical
playing (at least not in the same way) but I believe could make
interrupts approaching the same rate as someone playing a guitar.
> Finally, it is widely accepted in cognitive sciences that often one does
> already act before the 'act of action' gets in your consciousness.
Yes but I think there could be different levels of action (as above)
with different relationships with consciousness.
> This might be quite relevant for playing e.g. a guitar, whereas with
> programming, one definitely needs to be conscious about what you are
> writing (syntax restrictions etc.)
I don't think I am concious about language syntax when I am programming
any more often than when I am when I'm writing/speaking.
Aren't there syntax restrictions with the guitar too? Some actions
result in a crash (broken strings/wood), or no change to the output, or
are simply impossible to play under normal conditions. I may be
thinking wrongheadedly here though...
> So, there's definitely not only different types of abstraction, but also
> quite different levels of abstraction, computer programming being at one
> of the highest levels.
Yes, but perhaps it's an expression at a level of abstraction that is
not as visible with guitar playing but still present.
alex
[1] Jeff Pressing, Improvisation: methods and models, in Generative
Processes in Music, OUP 1987
Received on Tue Sep 05 2006 - 12:15:04 BST