From: Bert

Consider, for example, that when you pull a part off a machine, you
have no way of knowing what the original dimensions of the part were

it doesn't take much of a shift in the
agent/machine boundary to include the complete part specifications
with the agent, so replication of the plans becomes unnecessary.

Wow. That's really cool and a good application of Von Neumann's view. I'll have
to give you credit for that at some time.

So the question on the floor, drawn to our attention by Gary, is:

Where can the agent/machine boundary be?

Just anywhere? I think not.

In my view, agent and machine are one. We ignore the agent, as each of us is
one.

One idea that comes to mind is a self-documenting machine part: text engraved
on the part specifying tolerances, torques, fits, etc. We do this in the
experimental shop frequently, along with selective assembly and marking for
disassembly. As bearings are numbered to a catalog, parts can be labeled for
as-needed replication, the alternative being a very heavy inventory of
otherwise useless parts.

Gary, Ned, and Bert, give me a moment to collect my thoughts on the rest.

I'm sorry about the sarcasm.



Yours,

Doug Goncz (at aol dot com)
Replikon Research, Seven Corners, VA

1200+ original posts at:
http://groups.google.com/groups?as_e...uthors=dgoncz@