At 7:49 PM +0000 3/8/03, The Long and Winding Road wrote: >""Priscilla Oppenheimer"" wrote in message >news:[EMAIL PROTECTED] >> Wow. Good thing Paul didn't really bring down Group Study and make it so >we >> couldn't have this important discussion! :-) He said it would be down for >> maintenance, but I didn't notice any downtime, (not that I spent the whole >> night checking!) Thank-you Paul, either way, for all the work you do to >let >> us participate in these conversations. > >mega dittos > >> >> Seriously, have you ever noticed that some people see similarities in two >> things and jump to the conclustion that the two things are the same or can >> be put in the same category? We see that a lot of that on this list.... I >> guess it's because there's not enough info on how to categorize stuff like >> DV versus LS protocols.
One of the ironies is that many people don't understand routing protocol theory deeply enough to realize there is a very simple distinction: the updates of DV protocols are cumulative distance vectors of routes, and the updates of LS protocols are local states of links (links according to Dijkstra). Things like neighbor relationships, reliable transfer, etc., are all extremely secondary to these core concepts. ARFARFARFyapyapARFARFARF broadcasts tend to be DV. Mrrow? query with Mrrrph response are much closer to negotiated neighbor synchronization. My experimental data of 2-3 cats licking my face at the computer does not suggest antisocial behavior, nor does Mr. Clark sleeping against my side or in my armpit, Rhonda keeping my head warm, and Ding on top. In fairness, Mr. Clark looks a bit like a bulldog that happens to meow. > >first of all, we all like clear distinctions. something is either this or >that. ever notice there are two kinds of people: the kind who believe there >are two kinds of people, and the kind who do not? ;-> > >secondly, all of us who are not experts tend to fall back to cliches and >authoritative texts for our answers. we see this regularly in the OSI versus >Cisco's version of OSI, and we certainly see it in Cisco's characterization >of EIGRP and a"hybrid" protocol It would be nice if Cisco would characterize OSI to the extent that OSI architects have done. There was no accident that the ISO people found that the 1984 document, ISO 7490 OSI Reference Model, was incomplete. Even within the ISO camp, the basic document was supplemented with four annexes (e.g., connectionless communications, management), and an assortment of technical reports and other supporting documents (Internal Organization of the Network Layer, TR10000 Functional Profiles, the revised object-oriented focus on the upper layers, mostly seen in X.400-1988, etc.). Even beyond ISO, we have the complementary model refinement in B-ISDN, IEEE 802, etc., plus relevant architectural work in the IETF (e.g., GMPLS). Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64853&t=64853 -------------------------------------------------- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
