10-02-2000, 05:16 AM
Dear Mr. Lloyd,
My name is Nina Katchadourian, and I'm an artist living and working in
Brooklyn, New York (although I was born and raised in the Bay Area!). I've
been on the web researching Morse Code, where I found your site. I read
with great interest about QRZ and was struck by the fantastic spirit of the
project. Since you struck me as both very knowledgeable and friendly, I am
contacting you to see if you might have any input or advice on a rather odd
problem I'm trying to solve. I hope you might have a few minutes to read my
email and offer any suggestions or advice.
I've had a long-standing interest in language and translation, and my art
work often touches on these subjects. Years ago, while in graduate school,
I took a class on Morse Code, which has always interested me. My art spans
many mediums, among them sound and sculpture, and a new project that I'm
working on integrates all these things.
The idea, in a nutshell, is this: I'd want to build a machine that will
take sound of popping corn, and translate it, via Morse Code, into
language. A computer-generated voice will then read out the resulting
sounds or words. I imagine using a very typical popcorn vendor's cart, the
type you might see at a circus for example, which would then be hooked up
to various microphones and computers.
As I see it, there are two phases to this translation process: first,
translating from the sound of the popping corn into dits and dahs, and
second, translating the code into letters and words that can be read by a
computer and "spoken." The second phase seems relatively simple, but the
first part requires quite a bit of thought and work.
Popcorn only pops in short sounds, of course. I believe that, since it's
possible to "knock out" code, one could determine what a dit is and what a
dah is by the amount of space that follows a sound. One would also need to
determine the spaces between letters, and between words. I'm not sure how
to deal with the fact that popcorn "accelerates" as it heats up and begins
to pop faster.
"Meaning" is an interesting question with this project. There is a high
likelihood of getting a "monkey at the typewriter" effect. That might be
OK, but on the other hand, there are varying amounts of "bias" that could
be built in to the way sounds are interpreted. For example, a computer
might be linked to a dictionary/spell check function so that it hears the
sounds and "leans" towards making meaningful combinations of letters
whenever possible. If the popping has generated a D and an O, and the third
sound could be heard as either a K (-.-) or a G (--.), it will pick the G
to make "DOG." Or, perhaps the computer is set to hear vowels whenever
possible to increase the chances of meaningful words being generated. The
degree of accuracy in this project is something that I have to make
decisions about, but which is also partially determined by what's feasible.
I exhibit with Debs & Co. Gallery here in New York, and I will exhibit this
project there at my next show which is in early January 2001. This project
is going to be exhibited next year in San Francisco at Catharine Clark
Gallery, my representatives on the west coast (in fact, I have a sound
piece up there at the moment!). The project is also likely to travel. All
of this is happening shockingly soon, and the more information I can gather
now, the better it would be!
I don't do any computer programming myself (gee, can you tell?) and I'm
obviously going to have to find someone who can do that part of this
project for me. Since my background isn't in this field, it's hard for me
to gauage how difficult ANY of this is. At this point, I'm collecting
opinions from far and wide about different ways to solve this problem. If
you have any ideas, feedback or thoughts about the project, or if can point
me to someone who could help (or even DO the programming!) I would be
extremely, extremely appreciative.
Thanks very much for your time.
Sincerely,
Nina Katchadourian
Nina@immaterial.net
My name is Nina Katchadourian, and I'm an artist living and working in
Brooklyn, New York (although I was born and raised in the Bay Area!). I've
been on the web researching Morse Code, where I found your site. I read
with great interest about QRZ and was struck by the fantastic spirit of the
project. Since you struck me as both very knowledgeable and friendly, I am
contacting you to see if you might have any input or advice on a rather odd
problem I'm trying to solve. I hope you might have a few minutes to read my
email and offer any suggestions or advice.
I've had a long-standing interest in language and translation, and my art
work often touches on these subjects. Years ago, while in graduate school,
I took a class on Morse Code, which has always interested me. My art spans
many mediums, among them sound and sculpture, and a new project that I'm
working on integrates all these things.
The idea, in a nutshell, is this: I'd want to build a machine that will
take sound of popping corn, and translate it, via Morse Code, into
language. A computer-generated voice will then read out the resulting
sounds or words. I imagine using a very typical popcorn vendor's cart, the
type you might see at a circus for example, which would then be hooked up
to various microphones and computers.
As I see it, there are two phases to this translation process: first,
translating from the sound of the popping corn into dits and dahs, and
second, translating the code into letters and words that can be read by a
computer and "spoken." The second phase seems relatively simple, but the
first part requires quite a bit of thought and work.
Popcorn only pops in short sounds, of course. I believe that, since it's
possible to "knock out" code, one could determine what a dit is and what a
dah is by the amount of space that follows a sound. One would also need to
determine the spaces between letters, and between words. I'm not sure how
to deal with the fact that popcorn "accelerates" as it heats up and begins
to pop faster.
"Meaning" is an interesting question with this project. There is a high
likelihood of getting a "monkey at the typewriter" effect. That might be
OK, but on the other hand, there are varying amounts of "bias" that could
be built in to the way sounds are interpreted. For example, a computer
might be linked to a dictionary/spell check function so that it hears the
sounds and "leans" towards making meaningful combinations of letters
whenever possible. If the popping has generated a D and an O, and the third
sound could be heard as either a K (-.-) or a G (--.), it will pick the G
to make "DOG." Or, perhaps the computer is set to hear vowels whenever
possible to increase the chances of meaningful words being generated. The
degree of accuracy in this project is something that I have to make
decisions about, but which is also partially determined by what's feasible.
I exhibit with Debs & Co. Gallery here in New York, and I will exhibit this
project there at my next show which is in early January 2001. This project
is going to be exhibited next year in San Francisco at Catharine Clark
Gallery, my representatives on the west coast (in fact, I have a sound
piece up there at the moment!). The project is also likely to travel. All
of this is happening shockingly soon, and the more information I can gather
now, the better it would be!
I don't do any computer programming myself (gee, can you tell?) and I'm
obviously going to have to find someone who can do that part of this
project for me. Since my background isn't in this field, it's hard for me
to gauage how difficult ANY of this is. At this point, I'm collecting
opinions from far and wide about different ways to solve this problem. If
you have any ideas, feedback or thoughts about the project, or if can point
me to someone who could help (or even DO the programming!) I would be
extremely, extremely appreciative.
Thanks very much for your time.
Sincerely,
Nina Katchadourian
Nina@immaterial.net