From: Jeff Bone (jbone@jump.net)
Date: Sun Sep 10 2000 - 12:46:18 PDT
> Because they aren't best-of-breed. None of those ever were. The problem
> is that many people don't know how to value technology, not that the
> winners have somehow lesser value. All designs have trade-offs.
I think that's a little too facile. The transputer managed to have too many
great things in it that people didn't understand how to use at the time: it
was the earliest mass-produced RISC chip *and* it had addressable channels.
Both things showed up and were well-used in later chips, the latter in the
Intel iSeries chips. The problem was the market adoption curve, not anything
technical. But it's still a measurably better architecture in many respects
than iX60 etc.
Colecovision was best-of-breed at the time of introduction in both
resolution, on-screen color, and graphics operations per cycle.
NeXT was too much, too soon, at too high a price point. Etc.
The problem in all the examples I mentioned wasn't necessarily poor technical
trade-offs, it was more a general problem of "the innovator's dilemma." The
market only absorbs innovation incrementally. It's never a good business
idea to push the envelope too far ahead of the mainstream; building a web
browser, e-mail client, and protocol stuff for PDAs in 1994, for example:
bad idea. ;-)
jb
This archive was generated by hypermail 2b29 : Sun Sep 10 2000 - 13:09:18 PDT