RE: [DNA] RE: draft-ietf-dna-link-information-03

[Bernard Aboba <bernard_aboba@hotmail.com>]

I didn't realize this document was trying to take 802.21 into account.   I 
had assumed that it was targeting a relatively short-term completion date, 
so that it was not feasible to wait until 802.11r completed sponsor ballot 
later this year. But given that the gears of the IETF grind slowly indeed, 
perhaps there is little reason not to add new text.

>From: David Hunter <hunter@timefactor.com>
>To: Bernard Aboba <bernard_aboba@hotmail.com>,alper.yegin@yegin.org, 
>jari.arkko@piuha.net
>CC: Greg.Daley@eng.monash.edu.au,Suresh.Krishnan@ericsson.com, 
>gregdaley@ieee.org,dna@eng.monash.edu.au,hunter@timefactor.com
>Subject: RE: [DNA] RE: draft-ietf-dna-link-information-03
>Date: Thu, 25 May 2006 15:58:17 -0700
>
>Agree with Bernard on the antecedents, but not on the conclusions.  Bad 
>news:  to work with 802.11, 802.21 is going to have to accommodate 802.11r 
>(just a fact of the market).   Good news:  802.11r is much further along 
>than 802.21.  *If* no major glitches happen, 802.11r will be complete and 
>passed through as a standard by the time 802.21 hits sponsor ballot.  So 
>it's pretty safe to include 802.11r accommodations in a document that is 
>also trying to take 802.21 into account.
>
>Hunter
>
>At 11:14 PM 5/24/2006, Bernard Aboba wrote:
>>>Does the document provide any guidance on what the link up/down
>>>events should be for preauthentication or fast handoff (802.11r)
>>>situations? Or is it too early to specify these?
>>
>>EAP pre-authentication can only occur if the host is already connected 
>>(e.g. link is in the "up" state), so this is not relevant to DNA.  IEEE 
>>802.11r is a somewhat complicated because initial authentication uses the 
>>802.11i model (e.g. no IP connectivity prior to post-EAP handshake 
>>completion), but handoffs are more like the original 802.11 model ("link 
>>up" on completion of authenticated re-association).  Given that 11r isn't 
>>done yet, it is probably best not to try to cover it in this document.
>
>