Re: process vs. data

[Kenneth Lu <kenlu@MIT.EDU>]

At 10:39 -0800 2/2/01, Chris Crawford wrote:
>In evolutionary
>terms, data-intensive approaches are a dead end because they don't help us
>understand the deeper issues. Sure, you can memorize 8 guzzinta 40, but what
>happens when you need to know the value of 8.1743291 guzzinta 42.1917446?

I agree. I think that right now, when we talk about a "data-oriented 
approach", what we're really talking about is an approach where the 
"process" is in the mind of a human author instead of a computer.. 
where we use this delightful thing we call "intuition"..

I think that researching the process will, as is the purpose of all 
cog sci work, help us to better understand how our own minds work.. 
and in the process, maybe help us to automate certain aspects of our 
stories.

Now, in the long run, when/if computer are smarter than us and 
stuff.. then maybe they'll become even better storytellers than human 
beings.  But in the meantime, I think we should look to see how we 
can use computers to take away some of the burden of the process.. 
and exercise ITS strengths (i.e. repetition and heavy calculation) in 
concert with our own (intuitive (i.e. complex-system-based) sense of 
what makes for a "good story").

So I think we should move, as has been mentioned, in to the middle 
from both ends, and see where we go!

-ToastyKen

-----------------------------------------------------------
| Kenneth Lu - kenlu@mit.edu - http://www.mit.edu/~kenlu/ |
-----------------------------------------------------------
| "Life is far too important to be taken seriously."      |
|                                                         |
|                                          -- Oscar Wilde |
-----------------------------------------------------------