James,
I am fine with both 1) and 2), although 2) seems to be a bit
redundant
redundant; IETF is supposed to do --I hope-- good
engineering.
I don't think we should go for 3) because
"must" is too strong. It
might be that something is missing or too complex in using
existing
protocols to provide the desired NetLMM functionality (e.g.
state
machines have to be modified, or we need too many new
options.)
--julien
On Tuesday 21 March 2006 21:15, James Kempf wrote:
> This note is to continue discussion on Issue #16 which
applies to
> the REQ draft. The issue is whether to add an explicit
design goal
> to require maximum reuse of existing protocols and
backward
> compatibility in the NETLMM protocol. The three
possible options
> discussed at the meeting are:
>
> 1) Not add any text in the REQ draft for such goal.
Leave it up to
> the DT and WG to use best engineering judgement where
reuse of
> existing protocols is appropriate and where new
protocols are
> needed, so the goal is implicit.
>
> 2) Make point 1) explicit as a goal in the REQ draft
and say that
> the DT and WG should consider existing protocols when
doing the
> design and only do new protocol work where necessary.
>
> 3) Add a goal to the REQ draft that says the DT and WG
must reuse
> existing protocols.
_______________________________________________
netlmm mailing list
netlmm ngnet.it
https://vesuvio.ipv6.cselt.it/mailman/listinfo/netlmm
|