Alt-BEAM Archive

Message #03628



To: beam beam@corp.sgi.com
From: Richard Piotter richfile@rconnect.com
Date: Mon, 24 May 1999 14:35:25 -0500
Subject: [alt-beam] Re: BEAM and machine evolution


I read this, and I can only agree. I ain't about to stop using
alternative designs (to BEAM that is), and if I ever get around to
learning how to program a microprocessor, my BEAM bots will gladly carry
one on it's back. I've seen video of Cog, and that is an impressive
robot! Imagine placing it's top on the Honda humanoid's bottom!!! That
could get interesting! Hehe! (:

JVernonM@aol.com wrote:
>
> In a message dated 5/24/99 12:08:45 PM Eastern Daylight Time, sbolt@xs4all.nl
> writes:
>
> > BEAMer knowledge has difficulty expanding, because of the curious
> > accent on tinkering as opposed to design. For instance, there is no
> > way to get the various 2-transistor SEs `right', because there are
> > fundamental problems with that design, which no amount of tinkering
> > can solve. Hence these supposedly simple devices will forever be
> > a source of much frustration.
> >
> Actually Steven, I think there is another more important hindrance to BEAM
> knowledge expansion. In my opinion BEAM is a very tightly wrapped ball of
> biases. Your suneaters are an excellent example. Suneaters solve many of the
> problems with the S.E.'s, but almost no one builds them. You offer kits, but
> I would say you haven't sold many. You are not Mark T. or Dave H. so your not
> really BEAM. Even if you are, you aren't. The knee jerk reaction in the BEAM
> community to different, and even better ideas is to pounce with ferocity on
> the turf invader. Even if the ideas have merit, they are rejected as non
> BEAM, impossible, unimportant, or unnecessary. BEAM is a technology that
> revolves around the opinions of a small minority of people. They, through the
> use of certain bias, really control the flow of acceptance. Particularly on
> this list. Anything that deviates from the norm, even if sound and workable
> hasn't got a prayer if not accepted by that group. That's why people fiddle
> endlessly with FLED based poppers that barely work and completely ignore
> better options like the suneater. Evolution? I'm beginning to wonder. If, as
> Tilden suggests, robotics evolution is in us, then I'm beginning to wonder
> about it's viability. We don't accept better ideas unless they come from the
> perceived correct source. We shun better circuits and designs unless they are
> offered or accepted by the hierarchy. In truth, I'm beginning to think BEAM
> can't evolve. After all, BEAM seems to be simply a Tilden emulator. Most of
> us just want to copy Mark's designs, or Dave's designs, and pull them out
> every now and then to impress friends. I'm not so sure this is the way to
> robotic evolution. Perhaps Dave is right, BEAM is no more than a hobby meant
> to lead one to graduating to greater things. But, then there's that CPU bias.
> I'm really beginning to wonder if there is, or ever will be a high end to
> BEAM. I'm sorry guys, but when I saw Stryder shuffling across the screen on
> PBS last week, I couldn't help but remember what Tilden said about Cog at
> MIT. I mean, Cog is a very advanced system. It can recognize faces and
> expressions, and respond in kind. That sounds like serious research to me.
> But all Tilden could muster was that all it can do is barely keep from
> beating itself to death. And this bias filters down to the BEAM community as
> CPU's suck. Don't even look at them. As I watched Stryder stumble over one
> bot it didn't know was there and finally bump into another Symet that
> triggered it's leg tactile sensor, I thought, "This thing can barely walk."
> As far as I can tell, it only lifts it's legs a couple of millimeters off the
> table top. It's not even really walking. But, from Tilden's point of view, it
> has more merit as a precursor to robotic evolution than Cog. That inside out
> bias will forever limit BEAM to the hobby work bench. This attitude filters
> down until it becomes BEAM law. I know I rant on these things to much, and
> this is the last one. I am getting nowhere trying to explain what I perceive
> as destructive and limiting attitudes. If someone is hell bent on stifling
> their own creativity, there's nothing that will change it. I'm beginning to
> think BEAM will forever be the crystal radio. Fun to play with, easy to
> build, and a learning experience. But, it eventually has little to do with
> the radio industry as a whole. It's a shame really.
> Jim

--


Richard Piotter
richfile@rconnect.com

The Richfiles Robotics & TI web page:
http://richfiles.calc.org

For the BEAM Robotics list:
BEAM Robotics Tek FAQ
http://people.ne.mediaone.net/bushbo/beam/FAQ.html

------------------------------------------------------------------------

eGroups.com home: http://www.egroups.com/group/alt-beam
http://www.egroups.com - Simplifying group communications



Home