Joe Touch wrote:
Architectures aren't fixed things; they can evolve, e.g., you can go
from one stable architecture to another, more capable one by deliberate
extension.
Surely, something you call architectures might evolve. So what?
The real heart of the Internet is an architecture.
Depending on what you call "architecture", may be or may be not.
The E2E principle and
narrow waist are consequences of that architecture.
Completely wrong.
Apparently, you haven't read the paper, which does not depend
on specific structure/architecture of the Internet with a
very old example of CTSS, reference to which was written in 1963,
and another old example of telephone networking, reference to
which was written in 1964.
As the paper is titled
END-TO-END ARGUMENTS IN SYSTEM DESIGN
and its abstract begins with the following statement:
This paper presents a design principle that helps guide
placement of functions among the modules of a distributed
computer system.
the argument is applicable to system design in general and,
especially, all the distributed computer systems including
preliminary and *EVOLVED* ones.
Moreover, the paper does not say anything about architecture
of networking (though the paper briefly mentions CPU architecture
of RISC).
So, while you can call something an end to end architecture and
say it can evolve, it has nothing to do with the applicability
of the principle, the end to end argument, to system designs
with or without the evolved architecture.
With the argument, it can be concluded that approaches presumed by
SEMI2015 is incomplete and incorrect.
Masataka Ohta