Greg Menke  <gregm-xyzpdq@toadmail.com> wrote:
+---------------
| Tim X <timx@spamto.devnul.com> writes:
| > Even relatively simple protocols, like http are a pain
| > when described within a formal spec. The problem is with trying to
| > balance clarity/brevity without ambiguity - something human language
| > is not great at. This is often made worse by the unfortunate reality
| > that often those who are very talented in a technical sense are often
| > somewhat less talented with written expression.
| 
| Got that right.  The IEEE 802.* specs will make your brains leak out
| your ears.  They approach write-ohly documentation but if you keep
| slogging away its possible to extract content from them.
+---------------

Whereas the Xerox-written [despite being a "DEC/Intel/Xerox" co-effort]
Ethernet 1.0 & 2.0 specs that pre-dated 802.3 were *EXTREMELY* readable,
and presented the same information three different ways: plain English,
Pascal-like pseudo-code, and engineering-style electrical specs. Yet for
all their clarity and redundancy, they were a *lot* shorter than 802.3.

Hint: The Ethernet specs weren't written by "A Standards Committee"...


-Rob

-----
Rob Warnock			<rpw3@rpw3.org>
627 26th Avenue			<URL:http://rpw3.org/>
San Mateo, CA 94403		(650)572-2607