Hubert Lelong wrote in a message to Mike Bilow:
 > I hope this is a joke, because I cannot imagine anyone
 > learning this from the standards documents.  In any
 HL> no it is not. I read Christian's book. But very often a
 HL> suplier/reseller says 'my products are full standard" and
 HL> you must find by yourself if it is true or if it is not.
 HL> then you neeed to know how the Protocol has to work, and
 HL> show that it is not true !
I misunderstood your message, then.  I thought you were looking to learn the 
subject from the standards documents.  Apparently, you really did want the 
standards documents, which is what I sent you.
 HL> for example, I had to show to ACC that their version 7 was
 HL> not compatible with their RIP protocol as described in the
 HL> RFC ! without the RFC (and frame capture, of course) they
 HL> would have go on saying that's the problem of routing (with
 HL> soft routers from Netware) was quite normal.
RIP is such a mess, with so many revisions and variations, that it would be 
hard to find any perfectly standard implementation.  I would not be surprised 
if you could even find standard implementations which were not interoperable!
 > Radia Perlman's more technical "Internetwork Routing."  OSPF 
 > and spanning tree algorithms generally are extremely complex.
 HL> I'm going to see if I can find this book in France.
Books are hard to find because, for whatever reason, most of the people who 
have been involved with routing protocols verge on the incoherent.  I used to 
translate for one of Radia's co-workers at meetings: he used so many acronyms 
that he was incomprehensible.  After he actually tried to pronounce "SNAcP" 
in the course of conversation, I realized that he needed a translator.
 
-- Mike
--- 
---------------