Re: minutes of RIPv2 meeting in Amsterdam

[Gerry Meyer <[e-mail]>]

>Gerry Meyer's Demand Routing proposal was discussed at length.  It was
>agreed that it performed a useful function.  However, Robert Ullman
>pointed out that it simulated many of the functions of TCP and that RAP
>used TCP.  Robert and Gerry will continue the discussion after Gerry
>has had an opportunity to read the RAP RFC.  The slides from Gerry's
>presentation are included in these Proceedings.

Like all other current routing protocols, RAP (experimental RFC 1476)
currently does not address the problem of running a routing protocol on
a switched network.  It is also IP-specific in its current form.
I feel RAP is a red-herring in this discussion.

The real issue is what is the most appropriate method to operate on demand
circuits with RIP protocol variants (IP, Novell, Apple etc):

(a) an appropriate unreliable service (UDP-IP, IPX etc) which is what we
    currently use and is proposed in the Internet Draft.

(b) over a single reliable transport (TCP, encapsulating Novell RIP etc),

(c) the 'appropriate' reliable transport (TCP for IP, NCP or SPX for Novell
    etc).

I think the 'best' method depends on where you are coming from.  I rule
out (c) immediately as requiring transports not otherwise required in a
multi-protocol router.

How I intend to proceed is to sketch through case (b) for myself to see
what functionality it can provide.   I will then mail a list of pros and
cons for the two methods for a discussion on the mailing list.

    Gerry