Document: draft-ietf-ospf-scalability-08
Reviewer: Spencer Dawkins
Date: June 20, 2004

The only "unreasonable" thing I saw in this draft was on page 5 -
after a description of the problems encountered bringing up too many
adjacencies simultaneously, there's a suggestion that a limit could be
reasonably configured - but the text points out that available
bandwidth is one of the parameters in determining a reasonable limit,
and I would think available bandwidth would be just the kind of thing
that would change rapidly when you ARE trying to bring up lots of
adjacencies simultaneously. I'm thinking that this limit either needs
to be dynamic or that more guidance in selecting a configured value is
required.

Other than that - the draft seems wobbly between two priorities and
three priorities ("please pick one - the current guidance seems likely
to confuse implementors"), but is clearly written and has few nits
(the usual 2026 boilerplate comment applies, for instance).

Spencer