Hi, Cormac,
Deep Blue had such a glaring advantage over Kasparov that I've always been
amazed that no one brings it up.
Before the famous chess match, programmers entered every documented game
that Kasparov had ever played into the computer's storage. The computer then
had the luxury of analyzing Kasparov's style, biases, and behavior in a
multitude of chess situations. Kasparov, of course, had no such insight into
the computer's tendencies. One must ask himself: What would have been the
outcome had Kasparov been privy to a 30-year history of Deep Blue's chess
skill?
Nonetheless, why are we finding robot operation so implausible? Robots are
already communicating and exchanging information in much more difficult
conditions than the 20-meter band during the SS. Read about cross-country
meteor-scatter communications someday or, heck, the "robot" operation that
it will take to deliver this e-mail message from me to you for two examples
of complex automation. The protocols we have set up in amateur radio
contests are rudimentary and don't deviate very much: nearly ideal
conditions for robots to flourish.
As has already been pointed out here, the real question is not "How?" or
"When?", but "Why?" Ham radio contesting by robots will not supplant
operation by humans on a wide scale because of the added layer of
abstraction --we by and large aren't homebrewers any more--, making it
probably intolerable for the avid contester. For the progammer? Maybe the
point would be precisely to turn the experience of contesting into an
experience of creating machines that behave in remarkably pseudo-human ways
and then sitting back and watching, only to tweak the algorithm a smidgen in
time for the next contest.
R,
Al W6LX
-----Original Message-----
From: Cormac J Gebruers [mailto:c.gebruers@oceanfree.net]
Sent: Wednesday, May 31, 2000 1:25 PM
To: cq-contest@contesting.com
Subject: Re: [CQ-Contest] Robot Op Challenge
Oh a short addendum,
Its perhaps disingenuous to refer to Deep Blue beating Gary Kasparov
(Chess) - I would argue that such a feat proves nothing. Deep Blue operates
by a simple "Brute Force" approach - try every possible move and pick the
best one. That works because a chess game is a closed set - grant it a very
big one but it is finite. A contest is an open set - the possibilities are
infinite - i.e. Deep Blue wouldn't be much good at it :o)
So pat yourselves on the back - there's a lot more to contesting than chess
:o)
73
Cormac EI4HQ
----- Original Message -----
From: Derek Wills <oo7@pan.as.utexas.edu>
To: <cq-contest@contesting.com>; <n4zr@contesting.com>
Sent: Wednesday, May 31, 2000 7:04 PM
Subject: Re: [CQ-Contest] Robot Op Challenge
>
> N4ZR said:
>
> Frankly, I'm skeptical. I'm sure that robots exist with a
> limited degree of autonomous QSO capability (after all N6TR's
> Z80 Op has been around for a decade or more), but I doubt that
> the technology has progressed to the point that it can outperform
> a skilled human operator, even on a single band and over a short
> period of time.
>
> It's just a matter of time - the world's #1 chess player has lost to
> a machine more than once! And contesting machines don't get tired...
>
> Derek aa5bt
>
>
> --
> CQ-Contest on WWW: http://www.contesting.com/_cq-contest/
> Administrative requests: cq-contest-REQUEST@contesting.com
>
>
--
CQ-Contest on WWW: http://www.contesting.com/_cq-contest/
Administrative requests: cq-contest-REQUEST@contesting.com
--
CQ-Contest on WWW: http://lists.contesting.com/_cq-contest/
Administrative requests: cq-contest-REQUEST@contesting.com
|