Joel M. Halpern allegedly wrote on 11/17/2009 3:02 PM:
> Yes.  That was the point of my comment at the mike in the session.
> I like some of the soltuions that do host based handling, in part
> because they allow for things like multipath TCP.

You're going to get this in any case, for free :-).  An endpoint cannot
assume that a network it is connected to has a particular technology
(such as GSE), so it needs to take care of its own multihoming/mobility
needs.  No matter what happens in routing, endpoints are going to move
toward true location/identification separation.  This relieves the
pressure on RRG and opens up opportunities.

> But, if we are going to do that, we are going to have to actually make
> renumbering a LOT easier.  We are going to have to address the issues in
> Brian Carpenter's draft.  I believe that some of the issues will
> themselves be made easier by the separation of identifiers from
> locators.  But it will take more work than that.

Endpoints are easier than sites.


swb
_______________________________________________
rrg mailing list
[email protected]
http://www.irtf.org/mailman/listinfo/rrg

Reply via email to