SUO: The Law of Standards (comment on TCP/IP)
At 08:49 2002-06-07 -0700, Douglas McDavid wrote:
>
>
> John --
>
> I think another good example that supports your thesis is the success of
> TCP/IP, in the face of elaborate, well-funded standards efforts in the
> networking area.
Uh, wasn't TCP and IP well-funded by DoD and other research organizations?
The distinction between TCP and, say, the ISO protocols (and other networking and network management standards) was the distinction between distributed administration perspectve vs. central administration perspective. For this application area, a distributed administration perspective was more attractive to the marketplace ... these were the main decision choices people faced in the mid 1980s.
Additionally, TCP/IP really took off when a useful API paradigm became available, i.e., Berkeley sockets, around 1983. Prior to Berkeley sockets, TCP wasn't significantly better than any other protocol. I guess the point here is: a useful programming paradigm is important to the success of this kind of technology.
-FF
_______________________________________________________________________
Frank Farance, Farance Inc. T: +1 212 486 4700 F: +1 212 759 1605
mailto:frank@farance.com http://farance.com
Standards, products, services for the Global Information Infrastructure