Alt-BEAM Archive
Message #02453
To: Bruce Robinson Bruce_Robinson@bc.sympatico.ca
From: Bob Shannon bshannon@tiac.net
Date: Sat, 17 Apr 1999 22:31:23 -0700
Subject: [alt-beam] Re: History of Beam
Bruce Robinson wrote:
>
> So, the genome idea has sparked a lot of discussion, and kind of
> branched out in all sorts of directions. I think I may have been the
> first to mention the history aspect of it, but Dave's comments did cause
> me to focus my thinking a little.
>
> A history would be interesting; and now's the time to do it -- while the
> field is still young an memories haven't frayed too much. But that isn't
> really what I need. What I haven't been able to find in all my surfing,
> is a really good, structured, INDEX. Now I'm not refering of a typical
> index as found in the back of a book -- I'm thinking more of a very
> structured index that can be got at in different ways. Perhaps it's
> closer to a table of contents. For example:
>
> A genealogical approach:
> 1.) Walkers
> 1.1) True walkers
> 1.1.1) Single-motor walkers
> 1.1.2) Two motor walkers
> ... etc.
> 1.2) Walker-like
> ...
> 2.) Rollers (wheeled)
> 3.) Crawlers (tracked)
> ... etc.
>
> And a power source approach:
> 1.) Solar
> 2.) Battery
> 3.) Steam (tongue firmly in cheek here).
>
> And a "circuit" approach:
> 1.) VSPANS (read the Tilden patent!)
> 1.1) Monocore
> 1.2) Bicore
> ... etc.
> 2.) CPU based (OK, maybe not)
> ... etc.
Bravo!!
This is very much like what I had in mind with the encyclopiedia
concept.
It sould cover both brain and body.
It should include CPU based designs.
When such exists, standards of performance and behavior sets would also
be
added to the basic framwork.
> The idea here is that ultimately you could enter the index (or whatever
> it is) from the point of view you were interested in. You could then
> work down through the list to find what you were interested in. Or you
> could "surf" a branch -- say you got really good with some simple solar
> powered devices, and you were looking for more challenging solar
> projects. Ultimately, these lists would point to various web pages
> related to the subject.
>
> I've found lots of Beam web sites organized around various types of
> structures, pointing to other related sites. But they are all organized
> differently. I suspect many of them were attempts by their authors to do
> precisely what I am proposing -- but they only go as far as the author's
> personal needs (naturally).
This sounds perfect to me. There are far more bots than innovative new
design ideas. I think a guide to seperate the wheat from the chaff
would
be a valuable guide while designing new robots.
> Of course, you don't need to restrict yourself to a "mechanistic"
> approach -- you could use the same mechanism to build a "history
> approach", a "genome approach", or whatever.
>
> Is this useful to anyone? I know I need something better than the 75
> Beam bookmarks I've already got. I never get them organized right.
> There's too many ways to organize them.
>
> Key points:
> - Consistant format, so it has the same look no matter who builds it.
> - Compatible with as many browsers as possible.
> - Feedback capability via forms: solicit missing links, gaps in the
> history/genealogy/genome, new web pages, other ways to organize things
> that we haven't thought of, etc.
> - Ultimately points to as many beam sources as possible.
>
> OK, there it is. In way more than two sentences (sorry, Dave).
>
> Regards,
> Bruce
This sounds good to me!
------------------------------------------------------------------------
eGroup home: http://www.eGroups.com/list/alt-beam
Free Web-based e-mail groups by eGroups.com
Home