Hi Curt,
On 4 March 2014 21:08, Curt Cloninger <[log in to unmask]> wrote:
> Just to play devil's advocate (and to tease out even more nuanced differences between code and text), i would argue (following Bakhtin), that reading any human language text is an utterance event. So in this sense, anyone who has ever written anything (code, a shopping list, a sonnet) is a future typist, recording their utterance into a time-shifted medium that awaits any number of future readings (utterances/run-time events) in any number of unforseen affective, real-time, historical contexts.
Yes, interesting thoughts. Also, I wrote that text last summer, and
you've responded to it as a kind of reply to your post from earlier
today, which kind of plays into your hands. :)
> And any live jazz sax performer is involved in similar experiential feedback loops that live coders experience.
I often play with free jazz musicians, so I hope so.
> Maybe a way at it would be to take up Paul's provocations regarding something like max/msp/jitter. So what are the qualitative differences between real-time creation and modulation of a max patcher to compse/perform audio (always feels more deleuzean to me) VS something like livecoding in a command line programming language (which i imagine as somehow more derridean)? Based on your experience livecoding, what would you say?
It's an interesting question that I've thought about a bit, e.g.
http://yaxu.org/textual-patching/
I think there's no real difference. In one you connect words by
drawing a line between them, and in the other you do it by just
putting words next to each other in the first place. Some claim
drawing the line is 'more visual' than typing the text, but if
adjacency isn't a visual property, I don't know what is. Max patches
certainly contain a lot of text, anyway.
Max does have a dataflow model of computation, but then
non-diagrammatic languages can do that well, for example ChucK. I
think Max probably feels better for some things because they fit the
dataflow model really well, and because the editor doesn't let you
make syntax errors, which is great for exploratory programming.
With the collaboration with Kate, I'm using my Texture language, which
is actually Haskell but with a strange editor where you put things
near each other, rather than adjacent to each other:
http://yaxu.org/colourful-texture/
It looks a bit like dataflow, but underneath everything is pure
functions of time.
Cheers
alex
|