Okay: I would like a scientist to weigh in on this. I recently pointed out my own belief that the Million Monkeys Theorem is flawed; "random" doesn't mean "every single thing ever would be produced". "Random" means "we can't predict what will happen in any way."
So to say that a million monkeys would eventually write the works of Shakespeare is to misunderstand "random."
And I went and read the posts that you linked to, and here's my BIGGER problem with this: If you define "coming up with chunks of text that match" as "producing the works of Shakespeare," you can claim this project works EVERY TIME.
The blogger says:
oday (2011-09-23) at 2:30 PST the monkeys successfully randomly recreated A Lover’s Complaint and The Tempest (2011-09-26). This is the first time a work of Shakespeare has actually been randomly reproduced. Furthermore, this is the largest work ever randomly reproduced. It is one small step for a monkey, one giant leap for virtual primates everywhere.
But as far as I can tell, the "monkeys" didn't recreate it in anything like a complete format: It sounds like you have to search through tons of text to find bits and pieces of Shakespearean works, then piece them together to match up to a script. But you could probably do that in any book of word searches or collection of crosswords puzzles. How big of a chunk do the monkeys have to come up with? A whole page? Or just two words strung together? I wasn't able to tell.
If it's just two words, or just a few words, this isn't a test at all. Here's the entire text of Romeo & Juliet:
This comment begins "I would like." In doing so, I recreated a portion of that play:
SAMPSON
I mean, an we be in choler, we'll draw.
MONTAGUE ... would as willingly give cure as know.
CAPULET ... like her most whose merit most shall be:
Does that say anything about anything at all?
I'm going to recopy this comment onto his blog and ask him to comment on that. I'm no scientist, but I do know what "random" means, and as a lawyer, I also know how to rig results to get where I wanted to go.
Briane, I agree there's a huge difference between writing the individual words of a play randomly vs. putting them in order. Hopefully that means computer monkeys won't be writing us out of jobs anytime soon.
2 comments:
Okay: I would like a scientist to weigh in on this. I recently pointed out my own belief that the Million Monkeys Theorem is flawed; "random" doesn't mean "every single thing ever would be produced". "Random" means "we can't predict what will happen in any way."
So to say that a million monkeys would eventually write the works of Shakespeare is to misunderstand "random."
And I went and read the posts that you linked to, and here's my BIGGER problem with this: If you define "coming up with chunks of text that match" as "producing the works of Shakespeare," you can claim this project works EVERY TIME.
The blogger says:
oday (2011-09-23) at 2:30 PST the monkeys successfully randomly recreated A Lover’s Complaint and The Tempest (2011-09-26). This is the first time a work of Shakespeare has actually been randomly reproduced. Furthermore, this is the largest work ever randomly reproduced. It is one small step for a monkey, one giant leap for virtual primates everywhere.
But as far as I can tell, the "monkeys" didn't recreate it in anything like a complete format: It sounds like you have to search through tons of text to find bits and pieces of Shakespearean works, then piece them together to match up to a script. But you could probably do that in any book of word searches or collection of crosswords puzzles. How big of a chunk do the monkeys have to come up with? A whole page? Or just two words strung together? I wasn't able to tell.
If it's just two words, or just a few words, this isn't a test at all. Here's the entire text of Romeo & Juliet:
http://shakespeare.mit.edu/romeo_juliet/full.html.
This comment begins "I would like." In doing so, I recreated a portion of that play:
SAMPSON
I mean, an we be in choler, we'll draw.
MONTAGUE
...
would as willingly give cure as know.
CAPULET
... like her most whose merit most shall be:
Does that say anything about anything at all?
I'm going to recopy this comment onto his blog and ask him to comment on that. I'm no scientist, but I do know what "random" means, and as a lawyer, I also know how to rig results to get where I wanted to go.
Briane, I agree there's a huge difference between writing the individual words of a play randomly vs. putting them in order. Hopefully that means computer monkeys won't be writing us out of jobs anytime soon.
Post a Comment