[thelist] Re: philosophical or possibly somewhat OT

Luther, Ron Ron.Luther at hp.com
Mon Mar 24 14:50:53 CST 2003


Carol Stein asked:

>>To return to the real topic: Can anyone think of something we can do with
>>an RIA appropriately?
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

Hi Carol,

I'll bite.

Although I've always had an interest in AI, I believe I'm in agreement with 
the 'experts' that it takes more of a simulated personality than "intelligence", per se, to pass a Turing test.  

And although those kinds of creations can have many uses - I think I 
dread most of them: an automated insurance claims machine with a 'bad 
attitude' calculated and 'tuned' to minimize claims, an automated IRS 
interrogator, automated 'friendly' telemarketing 'agents', a pentium-based 
dial-a-porn service, etc. ... Bleah!

Furthermore, (as long as I'm kind of ranting), no matter how many heuristic 
algorithms you program - an 'expert system' is still a tool and it's 
"recommended advice" needs to be filtered by a wetware being.  ;-)

So where do I think RIA has a positive future?

I'd point to military and exploration opportunities ... say we drop a few 
thousand mechanical "ants" on Mars and leave them there for a few decades 
to gather surface data ... and collect the information later. I think that's 
a good use of the technology.

I'd also like to see an abstract math implementation.  I think we could 
develop a 'predicate calculus' for describing math elements, program 
the 'rules' [like contra positive] for proving a theorem ... and let the box 
crunch away at a high rate of dedicated sleepless speed ... printing out 
results that meet some preset criterion for 'interestingness'.


Anyway, that's a start on my 2¢,

RonL.



More information about the thelist mailing list