On Aug 18, 1:42*pm, nick hull wrote:
In article ,
*"Leo Lichtman" wrote:

(clip) I have an email out to sales at
Smithy asking for a part count and
pricing for the set of replacement parts needed to assemble a copy of
my Super Shop.
^^^^^^^^^^^^^^^^^^^^^^^^
Since this is a conceptual exercise, how about this option: *Order another
Supershop, and disassemble it. *You will then have all the parts required to
assemble a Supershop. *How is this different from disassembling your own
Supershop and then reassembling it? *Since this is no different from
disassembling my car and then reassembling it, could I say that my car is
capable of reproducing itself?

The difference is that the sum of the parts costs MUCH more than the
same parts assembled.

Yes, and that establishes one of several metrics by which self-
reproducing machine tool designs can be compared: If there are N parts
in a machine tool, and they cost m times the assembled price when
purchased as a set, then m is relevant. Also, if there are M distinct
parts in the set of N parts, then the ratio M/N is relevant. I haven't
named these ratios. but they are relevant. I feel like M/N should be
called the "distinctness fraction", with a value of 1 for an assembly
of distinct parts, and lower non-zero values for more evolved designs.
m might be called the "assembly disincentive".

Widespread adoption of a self-reproducing machine tool design seems
like it will have the effect of reducing m. On site, at the shop
floor, at the factory that builds the machien tool the way we do
things now, m 1. Out in the consumer market, m 1. Kinda makes me
want to be where m=1. Where would that be?

Probably in my home shop, or yours.

Doug (who is just tickled to drop by here again)