From: Adrian Bowyer (A.Bowyer@bath.ac.uk)
Date: Fri Mar 25 2005 - 00:46:00 EET
Quoting Bathsheba Grossman <email@example.com>:
> > Hmmm. Doesn't human engineering as a whole also constitute a
> > machine?
> OK, I'll buy that for a quarter.
Damn - only a quarter. Does that mean I lose 75 cents from the original bet?
Or am I now owed $1.25?
> It certainly did have vast
> consequences. But I only buy it if that system is taken to include
> DNA-based replicants -- if you take them out of the system at any
> point, it immediately stops perpetuating.
Sure - but my machine will self-copy but _not_ self-assemble. That's the whole
point (see the web site). It's much easier to make a pile of parts carefully
designed and labeled lock-and-key like so they only assemble one way and so any
(moderately competent) person can clip/screw them together. People will do that
if they want them - they are in the loop. If you want to strike grand
biological analogies (and I do little else) the machines and the people form a
symbiotic pair: both gain from the relationship.
> So who's this Matt Moses guy? I'm not finding anyone by that name at
Sorry - he's now at General Dynamics Land Systems. But see
There was also a gorgeous self-replicating machine built by Lionel Penrose
(father of Roger, who also did some of the work even though he was at school at
the time) back in the 1950s out of wooden blocks. For lots of universal
constructor stuff check out:
This archive was generated by hypermail 2.1.7 : Mon Jan 02 2006 - 08:09:08 EET