On Jul 18, 2009, at 7:23 AM, Joel M. Halpern wrote:
I think that this is at least misleading.
From a pure routing architecture perspective, the statement that
IPv6 started with (thou shalt always use provider address) is
sufficient.
However, once we acknowledge that issues like renumbering and the
realities of building networks are relevant, we have to look beyond
just our playpen and pay attention to adjacent activities and our
affects on them. Similarly, when evaluating map-and-encaps ideas,
we should be aware of the interaction of those ideas and other
activities like multi-path TCP, even though multi-TCP is clearly not
"routing".
Joel, I fully agree with the above statements (except the one about
misleading:)
e.g. taking multipath-tcp as an example: I thought this is covered
under my 3rd statement below:
We need to have a good understanding about the interplay
between addresses and identifiers
If we claim to have solved something with too small a focus, we
won't have anything usable.
agree.
Yours,
Joel
Lixia Zhang wrote:
...
But I would like to step up a level and repeat what I said earlier:
1/ I do not think it is in our charter to define how many
identifiers we ought to have, or what they ought to be.
2/ Our job is to figureout scalable routing architecture.
3/ We need to have a good understanding about the interplay between
addresses and identifiers, no less and also no more.
...
_______________________________________________
rrg mailing list
[email protected]
http://www.irtf.org/mailman/listinfo/rrg
