From: Yash (yashk2000@yahoo.com)
Date: Sat Jan 05 2002 - 03:29:15 MST
Herm,
Do you know Darell Huff's "How to lie with Statistics"?
Anyway, here's an interesting exercise for you:
Calculate the upper threshold probability that a certain language has a
man-made code built within it. That is, all texts in natural languages could
give you some recognisable patterns like PI and others to a certain extent.
If you can get the high threshold (which is near impossible as you would
only be able to say that this language conforms to 90% of confidence of the
target characteristic), then you could use this threshold to determine
whether a certain language was consciously built to incorporate such codes.
i.e. After your research, you get: Texts (of a certain size) in english
have a .90 probability of yielding mathematicals constants to 40 decimal
places.
Then you get Texts in language X, where the same probability is .99999999.
You'd be bound to consider this as indicative of the language being
constructed to do just that.
It's only this measure that should you really look for, not the supposedly
precise program thatn you wrote which fooled some people (but not me).
Just thought I'd mention it as you claimed I don't understand statistics.
Yash.
_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com
This archive was generated by hypermail 2.1.5 : Wed Sep 25 2002 - 13:28:38 MDT