Re: word source (pinker, chomsky, etc...)

>at times i feel stuck in that mode, and i'm still relatively new to
>much of this (i've only been writing code for little over a year,
>been working in flash and actionscript less…)


well ideal then. don't get in the habit like the rest of us. We
need a generation with a new take. Maybe somebody needed
demo-programming at one time but now that's over. You will show us
new approaches.

specific to your idea though:
sounds like a huge, ambitious undertaking regardless of the computer
language used. In fact, none I know of offer the slightest shortcuts
and you are just as well to use ActionScripting (which, for this, is
essentially JavaScript) And a brief glance at the code you posted,
you definitely know all you need for this task. Make a function,
access an array.

But sit down and do it. keep at it until you are impressed with the
results. Screw the obstacles. We need things to overcome to get to a
higher view. Be happy when they come find you.


But look forward to hearing what you make of this stuff.

judson


~~~~~~~~~~~~~~~~~~~~~~~~~~

PLASMA STUDII
http://plasmastudii.org
223 E 10th Street
PMB 130
New York, NY 10003

Hi Lewis
I thought a little about a similar problem a while
back -
I kind of had in mind the idea of a 'Shakespeare
Machine' which would spew out endless tracts of
wierdly almost shakespeare like stuff. A bit like the
infinite number of monkeys with typewriters.
It struck me that it might be quite fruitful to adopt
a statistical approach, that is to analyze a block of
text in a particular language formally simply ( or
perhaps not so simply) in terms of the frequency of
occurence of symbols and of prohibited and permitted
symbol combinations .I believe this is how textual
analysis by computer is conducted for literary
purposes.
( the basic symbol for our purposes being either the
letter or the word - the question of the actual
pronounciation of words in real life is of a massively
higher degree of complexity)
I think that this approach would then be much more
amenable to coding ( of course this wouldn't actually
produce english or italian or whatever if you were
working at the level of the letter, but something
hopefully disturbingly like it and the bigger your
original corpus and the more succint your statistical
distillation the nearer in kind to the original any
result)
If you were analyzing with the word as the basic unit
you'd get nonsense but interesting and evocative
nonsense.
I never did anything about it because life's too short
in my view and it would require serious amounts of
time and commitment to do properly.
It would be interesting to see someone give it a go.
regards
michael
— Lewis LaCook <[email protected]> wrote:
> >
> um, yes…
> i agree, actually===of course, i could hide behind
> the idea that this is nothing like a finished
> work===whatever a finished work is===but i simply
> must confess that i could not even BEGIN to code in
> a way that would make the computer recognize
> grammatical rules, much less phonetic
> custom===though someday i'd love to do it===
> i definitely agree w/ your comment on "more than the
> demo features"===at times i feel stuck in that mode,
> and i'm still relatively new to much of this (i've
> only been writing code for little over a year, been
> working in flash and actionscript less…)…
> hmmm…i'll have to reply to this when i'm in a more
> serious mode…this is thought-provoking!
> bliss
> l
>
> + your mama don't dance and your daddy don't rock &
> roll
> -> Rhizome.org
> -> post: [email protected]
> -> questions: [email protected]
> -> subscribe/unsubscribe:
> http://rhizome.org/preferences/subscribe.rhiz
> -> give: http://rhizome.org/support
> +
> Subscribers to Rhizome are subject to the terms set
> out in the
> Membership Agreement available online at
http://rhizome.org/info/29.php3


__________________________________________________
Do You Yahoo!?
Sign up for SBC Yahoo! Dial - First Month Free
http://sbc.yahoo.com