Re: More on new work item

Here's what I'd like to do:

Jonathan raised some good questions.  I'm not sure I agree with some
of his assumptions, but the question remains: is there an expectation
that someone would actually want to use DHCP to configure middlebox
addresses into endpoints, or is this an apparently reasonable idea that
won't be used in practice?

If we can't foresee a reasonable expectation that the work would have
real-world utility we should probably let it drop, and do so quickly.
If we do think the work is useful we should move it along as rapidly
as would be reasonable.  The main thing is that I'd like to avoid 
dinking
around with this, and I hope we can reach resolution quickly.  So, if
anybody wants to argue in favor of the applicability of the work, now's
the time.

Thanks,

Melinda