> A serious case can be made that the 7 layer model did more
> harm than good by helping the congenitally or willfully clue
> challenged pretend to be knowlegable and authoritative.
> Without the simplistic view of networking required by the 7
> layer model, the experts might have been forced to find other
> areas to plow.
I concur. The 7 layer model had its most harmful impact where it is most used, which
is in education. By the late 80's, a pretty standard curriculum was to have a 7
session networking course, with one session per layer. This was quite convenient for
the naïve teaching assistant who had a general computer science background but no
specialisation in networking. It was also quite wrong, since it manufactured
generations of students who believed that the OSI layering was somehow logical, when
in fact it was the result of accomodating diverse pre-existing systems, such as X.25
or teletext.
The basic failure was to "teach the standards" rather than "teach the technology." If
you teach the technology, then you deal with algorithms and scaling issues, and there
is plenty to teach and learn: retransmission algorithms and the use of timers, various
approaches to congestion control, name resolution, resource allocation algorithms,
marshalling and un-marshalling, synchronization and time, and many more. If you teach
the standard, then you content yourself with whatever a committee picked. Worse still,
in many case the committees' decisions were grounded in political compromise and
cloaked in technicalities. In many cases, the students got such a distorted view that
you litteraly need to make them "unlearn OSI" before they can do productive networking
work.
-- Christian Huitema
