Alt-BEAM Archive

Message #03950



To: beam@sgiblab.sgi.com
From: Jean auBois aubois@trail.com
Date: Sat, 29 May 1999 15:55:57 -0600
Subject: [alt-beam] Re: Unicore BeamAnt schematic!


At 03:48 PM 5/28/99 , you wrote:
>I bet you can't guess what I have. If you guessed that I have the
>schematic for a Unicore BeamAnt you get the grand prize!

In fact, this is one of my major BEAM sadnesses just now. (footnote: second
postscript, below) Mark kindly permitted me to make a xerox of the
schematic at least 1 1/2 years ago & also about that time allowed me to
copy the handout that was distributed at one of the Telluride Neuromorphic
Workshops. Both of those pieces of paper look pretty tattered because I've
read them so many times. I've sat on the bulk of that information ever
since -- originally, at that time, because he hadn't shown it to the small
gods yet (and requested that I keep my overly large mouth shut) and later
simply because I didn't feel I had permission. Even though he got rather
angry at me when he thought I was distributing "proprietary" information
about Lampbot, I wasn't -- the information has either been freely available
on the Web for -years- or was easily deduced given Suzanne Stills' papers &
visual inspection of Snakebot. Now, there are other, valid reasons for him
to be legitimately angry at me which I won't discuss here, but when it came
to "trade secrets" I believe I've done a good job of keeping quiet
(although I may be deluding myself about this).

However -- just for the sake of jollies, let us presume (well, it IS true,
but just for the sake of argument) that Mark is the leader in BEAM control
technology innovation (although Mr. Rigter is giving him a close chase.)
Although he has been -exceedingly- generous supporting isolated individuals
& the Los Alamos Workshop & the Telluride Workshop & the small gods
(whoever THEY might be) and so on and so forth, he hasn't been supporting
this bunch right here (yeah -- you -- the reader) as far as I can see.
Either we get the same exceedingly high-level ElectricTalk he uses to
market the BEAM concept (which isn't particularly applicable on a practical
basis) or silence or yet another promise that his book will come out some
day Real Soon Now so just wait. I can hardly remember ANY article
published directly by him on this list that describes the stuff he has
actually built in concrete terms. The little things in popular science
magazines and the incredibly esoteric articles published in certain
professional journals don't really count because the first contains the
same old same old and the latter doesn't give you enough/any information
about the Tilden GeeWhizRobot '97 to build one for yourself.

Of course, the usual rationale for not doing this is that he'll be deluged
with email & he'd certainly like to answer it (he is, in fact, very
generous) but he wouldn't have the time & that really doesn't set up a very
healthy situation for him. As such, this is understandable. Three hundred
messages asking for circuit details (and not just the old "which way do I
put in the resistor" routine) is more than anyone could bear and answering
even a portion of them would stop the rest of his work dead.

Is this truly an insoluble problem, though? For example, a moderately
simple solution to this might be: that he gets someone else (a person or
set of persons, probably some of US) to whom THAT mail is forwarded, as
automagically as can be done. I realize that email filters don't do the
same quality job of a human secretary, but I nonetheless find them useful.
Done properly, he would never see the distressingly large bulk of that mail
and only the truly interesting or incredibly arcane questions would filter
back to him.

A middle rationale for not doing this would have to do with the training of
the people who would have to deal with the deluge of mail created by the
above solution. That could be a real drag, but I'm pretty sure I can
discuss the BEAMAnt or even the newer SMT 2.0 Bicore boards intelligently
(and better once I really understand Rigter's and Wouter's explanation of
bicores better) -- and I'll bet that others on this list, providing they
were willing to COMMIT themselves could discuss Mark's various creations
with a minimum of his time & effort if they had been told about them in any
detail in the first place and were given permission to write about them
publicly. [I admit that my understanding of the Bicore boards is somewhat
dated, but it is to be hoped that the delta would be small.)

A less known rationale is an attempt to keep the information proprietary.
From the day his non-profit organization begins (or began, if it has
already happened), that organization can set policy on the dissemination of
information. Otherwise. . . well, I'm not sure of the ramifications here
-- legalities concerning "intellectual property" are still on exceedingly
shaky grounds -- for example, this is a _National_ Laboratory so doesn't
the work belong to the Nation (um, old inside joke. sorry). Perhaps
keeping the information proprietary is a good idea, perhaps it is not.
Perhaps the information simply can't be released because it is owned by
someone who won't or truly can't release it, perhaps it is public (at least
to the United States) property. I don't know, and even bringing up this
topic is frightening because I'm such an idiot in this domain.

To the best of my knowledge, Mark is operating in an utterly ethical manner
with regard to this stuff.

No matter what, given not only that Mark is the foremost innovator in the
field, that he has produced several years of robots, and that he doesn't
seem to be releasing any of that information to this list about them in a
practical manner... well, it just plain _hurts_. Once again, mebbe he
doesn't have to. However, if this list isn't the foremost body of human
beings interested in BEAM in the world, if it doesn't contain the most
ardent supporters (although financial support would be a Good Thing), then
what is it? Perhaps we don't -deserve- the information, but wouldn't it
vastly further the field if we had it?

In summary:

I request, I beg that the readership (yes, this means YOU again) discuss
with Mark how to break the information logjam. I don't pretend to have a
solution but then, what do I know . I'm incredibly frightened
about getting flamed out of my wits for posting this, or getting in
official trouble, or even hearing dead silence as a result.

I just wish I knew how we could get the most people working on the most
projects based on as much of the research that Mark has so successfully
produced as is possible.


p.s. Once again (just because I'm expecting this complaint,) yes I KNOW he
doesn't have to do anything he doesn't want to. I understand. Please, I
beg of you, don't climb all over my back about it.

p.p.s Ok, I understand that whether or not I feel hurt is probably not the
least bit important to you -- you aren't responsible for any of my feelings
and I hope I'm not coming off as a crybaby. If I am, I apologize in advance.

p.p.s. Perhaps BEAM isn't stuck (it feels a little stuck: yes, we've got
nifty new Rigter circuits and all, but how many people have built robots
with them? -- Mark has CERTAINLY built robots with HIS circuits) but it
still feels like it is at one of those points in a dynamic system when
things break wide open. Again, my feelings are not your problem. Your
mileage may vary. Please adjust your seatbelt. No smoking in aisles E, D,
and A.



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

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



Home