
www.Usenet.com
| <-- __Chronological__ --> | <-- __Thread__ --> |
On Fri, 7 Nov 2003 08:58:14 -0500, alck <[EMAIL PROTECTED]> wrote:
>
>OK then, "perfect" NLP will require a set of rules that predicts violations
>of rules....
>
>Saying that people 'violate' rules through sarcasm, irony, or what have you
>places blame on the subject and not the engineer. Sorry, wrong answer.
i don't think the blame is on the subject, or engineer, but simply the fact
that human language is incomplete, and always changing. i don't want to rehash
the statements already made in this thread (world knowledge, semantics, etc.)
so, just think about kurt godel's work, and that only applied to integer
number theory! my point is this, if integer number theory, a very formal and
unchanging language is incomplete, think about something as informal and
dynamic as human language.
that being said, however, i agree that we do need to keep researching, and
trying. it may be (and imho most certainly is) much more simple than we all
think. getting there though, may not be. =-)
to the journey!
peace,
cj
--
===============================================================================
Christopher Jon Miller Drink and dance and laugh and lie
Parallel Systems Engineer Love, the reeling midnight through
For tomorrow we shall die!
(But, alas, we never do.)
-- Dorothy Parker, "The Flaw in Paganism"
===============================================================================
| <-- __Chronological__ --> | <-- __Thread__ --> |