Gerhard Fiedler email (remove spam text)
Bob Blick wrote:
> How come market forces weren't enough to do such a simple thing as all
> car manufacturers agreeing on common data protocols?
Or end the high cost with the USA industry at large still not having moved
to metric (which it eventually will)? Or end the high cost of inter-bank
transactions being fed through the current system (which takes some four
days until a simple bank-to-bank ACH transfer can be considered confirmed,
and even then it's not possible to identify the sender at the receiver's
end or add something like an invoice number to the transaction)?
IMO that's because market forces have a relatively small horizon, both in
time and in spread. As Olin pointed out some time ago, for most companies
it's not economical to do something like this (adopting a common protocol,
adopting common measurements) as long as not everybody does it. And
everybody won't do it unless it's required by law. This creates local
optimums, with a relatively small range, and the direction towards these
local optimums may be contrary to the direction of larger scale optimums.
But companies rarely will take that move, especially not if the larger
optimum depends on other companies making the same move.
See also: www.piclist.com/techref/index.htm?key=better+designed
You must be a member of the
piclist mailing list
(not only a www.piclist.com member) to post to the