I'll rephrase "comparatively long";  what I intended was "long enough
so that we don't want to delay the current document for it."
 
I agree that the dual-MAC solution is important enough so that we
should push hard to make it work;  I also have a feeling that modifying
ARP (or replacing it) will have strong enough political overtones in
the IETF that it will be necessary to generalize the issue in order
to get the work done.
 
I also feel that we could draw on the expertise of others in the Internet
community by generalizing.  There are doubtlessly people who don't
particularly care about FDDI but do care strongly about ARP and
architecture issues.  This is a two edged sword, of course.
 
I think that the strongest interest in multi-rail LANs will come from
the FDDI community and I doubt that we will lose sight of the FDDI-
specific issue.