From: Danny Wyatt (danny@cs.washington.edu)
Date: Sun Oct 03 2004 - 23:50:20 PDT
The Design Philosophy of The DARPA Internet Protocols
David D. Clark
This paper is a brief historical review of the motivations behind
decisions made in the development of protocols for the Internet. It
explains the prioritized requirements for the Internet as it was first
defined and shows how trade-offs were weighed along those requirements
in order to come to the decisions that added up to (for the most part,
as the paper's focus suggests) a stateless datagram network. As such,
it does not have the usual results, assumptions, limitations, or
methodology (aside from, perhaps, historiography) to critique. It does
the job it set itself out to do, which is not a scientific one.
I found most interesting the way that a design decision for one
requirement harmoniously satisfied others. For example, the
"fate-sharing" approach to handling network failures meshes with the
decision to allow multiple transport services: both push functionality
out to the hosts. That "lightweight" design philosophy for the network
that emerges then satisfies both the reliability and generality
requirements.
I also found interesting the retrospective evaluations of which
decisions turned out to be more beneficial for reasons that did not
contribute to the decision in the first place. I'm curious to see more
such evaluations now that more than 15 years---in which the Internet saw
a somewhat more than modest increase in size---have passed since the
paper was first published.
This archive was generated by hypermail 2.1.6 : Sun Oct 03 2004 - 23:50:53 PDT