Re: proposed text for charter

[Jari Arkko <jari.arkko@piuha.net>]

Hi Erik,

> While the text is certainly more informative, I'm not sure this is
> a complete list of constraints, requirements, or goals.
> So is this a list of considerations (which isn't necessarily 
> complete), or is it a mixture of assumptions (such as multiple PA 
> addresses being used)?

Well, it was my list of most important constraints, requirements, and
goals. I acknowledge that it might not be complete. Do you have
something specific to add? Would you like to have a list of considerations,
or a full list of constraints/requirements/goals?

(Btw, I viewed PA addresses more as a constraint on the solution than an
assumption -- after all, if we could use PI addresses the solution would
be quite different. But my understanding was that this wouldn't be
possible for the type of target "market" that we are looking at.)

> > o  ULIDs are actual IP addresses such that existing applications can
> >   continue to work unchanged, and that application referrals work.
>
> The draft-ietf-multi6-app-refer is a bit more nuanced than just "work 
> unchanged". Based on what we know today, we can make unmodified 
> applications work as long as all the address pairs are working, but in 
> order for the referal-type cases to take advantage of all the address 
> pairs, the applications doing referals either need to use FQDNs, or 
> lists of addresses instead of a single address.

Yes, that's right. Do you have suggested text?

--Jari