modest suggestion for how to proceed

[Brian E Carpenter]

Harald Tveit Alvestrand wrote:
> 
> it seems clear that we've identified that there isn't just one core problem
> - so splitting up the task of dealing with them might be the Right Thing to
> do.

However, before we do that we need consensus on what the core problems
actually are. I'm suspending judgement until I see the next version of the
draft, but I'm not sure we are there yet.

I respectfully disagree with Keith. I don't want this work to fall into the
same sort of black hole as the multi6 WG (to take an example). I think we
need to start looking for targetted incremental solutions ASAP. Some of these
problems have been recognized for several years, so it's time to start
designing the fixes.

   Brian