[ipv6-dir] RE: Updated document

Pekka,

Wouldn't be correct to say that the IETF is working on the
identifier/locator split. I'm not
sure we need to say that it is a approach, but this is an area that the
IETF is working on.
I doubt we want the ITU-T making there own locator-split solutions!

Also, there are various MIPv4/MIPv6 dual stack approaches in the works
as well.   Various people
have talked about SCTP as a IPv4/IPv6 transition mechansims, and then
some people want to 'port'
SHIM6 to IPv4 as well ... 

I still like to think that IETF makes protocols and works on the
end-to-end architecture, and
other SDOs use the stuff we make, as they need them.

John

>-----Original Message-----
>From: ext Pekka Nikander [mailto:pekka.nikander@nomadiclab.com] 
>Sent: 05 January, 2006 16:02
>To: Kurt Erik Lindqvist
>Cc: ipv6-dir@ietf.org; Margaret Wasserman; IAB IAB; Loughney 
>John (Nokia-NRC/Helsinki); Brim Scott W; Bradner Scott; Leslie Daigle
>Subject: Re: Updated document
>
>> Please send comments!
>> <q.9-sg13-liasion-20060105.doc>
>>
>
>If we want to include speculative and forward looking text, I 
>propose the following for the "IPv4/IPv6 transition" bullet:
>
>---
>
>A potential future approach to IPv6/IPv4 interaction is to 
>implement the so called identifier/locator split in some way.  
>Among the various potential approaches being considered, the 
>approach used in the HIP working group 
>(draft-ietf-hip-arch-03.txt, in RFC Editor
>Queue) is among the more developed.  The parallel HIP research group  
>is studying the problems related to real life deployability of HIP.   
>If later decided to be adopted, HIP or some other 
>identifier/locator split approach may make the difference of 
>IPv4 and IPv6 irrelevant, allowing nodes to be mobile and 
>multi-homed between the two versions of IP, without any 
>IP-version related visibility to the applications.  [Technical 
>remark:  While the architectural approach taken by 
>identifier/locator split is different from the current 
>tunnelling and encapsulation approaches, at the actual 
>data-transport protocol level there is little if any 
>difference.  The differences lie in IP-layer state set up 
>between peer nodes and at the APIs available to the upper 
>layer protocols and applications.]
>
>---
>
>However, that text may be too speculative and/or something 
>that cannot be considered as a part of the IETF consensus.  I 
>am definitely biased here, and leave it for others to make the 
>evaluation.
>
>--Pekka
>
>

_______________________________________________
ipv6-dir mailing list
ipv6-dir@ietf.org
https://www1.ietf.org/mailman/listinfo/ipv6-dir