[fmc] Need for a technical discussion before the BoF (was RE: BoF session in Atlanta)

[<[e-mail]>]

Behcet,

See inline. 

Cheers,
Med 

>-----Message d'origine-----
>De : Behcet Sarikaya [mailto:[e-mail]] 
>Envoyé : mercredi 17 octobre 2012 05:51
>À : BOUCADAIR Mohamed OLNC/OLN
>Cc : [e-mail]; [e-mail]; JACQUENET 
>Christian OLNC/OLN; DUREL Sophie OLNC/OLN; KLAMM Frederic 
>OLNC/OLN; BINET David OLNC/OLN; SEITE Pierrick OLNC/OLN
>Objet : Re: [fmc] BoF session in Atlanta
>
>Hi Med, all,
>Thanks for your mail.
>Yes, I read your new draft,
>draft-boucadair-intarea-host-identifier-scenarios-00.txt, of course we
>know that UE identification is not specific to fmc. Who said so?

Med: You missed the point. The point is this issue is encountered in several use cases, the one called "fmc" is only one among others. 
* Why the IETF should focus only on this "fmc" case? 
* Wouldn't be more interesting to cover all UE identification use cases if there is a room for it? 
* If there are no other important technical issues, wouldn't be more pragmatic to focus on the UE identification technical issues?


>The reason we have been having this use case was because the request
>for IETF to be involved in fmc work has first been made during
>November 2011 3GPP-BBF FMC Workshop, Sophie would remember it very
>well, in the context of UE Identification use case which is in fact a
>well known problem with NATs and address sharing.
>
>Can I ask you how do you evaluate the chances of Intarea WG taking up
>this issue?

Med: This is not the question. We are not there at this stage. That draft is an input to the discussion in this list. I really think we need to have technical discussion in this list and more discussion on the issues before going asking for a BoF on behalf of people subscribed in this list. Going back to the minutes of the barbof organized in the last IETF meeting, there were a lot of questions about the issues and use cases. I don't remember reading in those minutes a GO for asking for a BoF.
 

 I have seen the reviews of your NAT Reveal draft. I think
>that Intarea people are very reluctant to work on this old problem
>related to mainly IPv4. I wish you good luck in this, maybe you can
>turn things around.

Med: We need to be fair here: intarea WG accepted to document the issues (RFC6269) and also to analyse HOST_ID solutions. Intarea WG is a candidate WG among others to host some specification work if there is a need for it. We need to convince if we want that to happen. This is the analysis work I asked for it to happen in this list and which was explicitly mentioned in http://tools.ietf.org/html/draft-schott-fmc-requirements-03#section-2. 

>
>fmc is a set of issues arising from 3GPP UEs accessing a fixed
>network. 3GPP and BBF will be busy developing standards for fmc at
>least next one if not two years. Afterwards, some next generation
>activity might be started, who knows? So fmc is here to stay.

Med: Why not working on a liaison from 3GPP-BBF? 

>
>I have been reading your mails and drafts recently and noticed that
>you mentioned almost in every mail or two, some issues related to fmc
>and also I have been seeing references to fmc in your drafts.
>
>So it is fair to say that fmc has made you more productive in IETF.
>Correct me if I am wrong, shouldn't you give some credit to Dirk and I
>for carrying the fmc flag in IETF and giving some inspiration to the
>researchers?
>
>Coming back to the UE identification use case, I am convinced more and
>more that we made a mistake in presenting UE Identification as the
>signature use case for fmc. In fact we can easily drop this use case
>and we will and can still have a lot to discuss on fmc in IETF.

Med:  Would it be possible to initiate a technical discussion on the other issues in the list? This will be helpful to prepare the BoF. 

 We
>will still keep our FT colleagues as co-authors if they wish so but we
>will give them a chance to drop out.

Med: I don't remember you provided a word in the requirements draft. We will see what to do with that draft.

>
>As you know, fmc BoF is scheduled already and we have no possibility
>of removing it from the agenda.
>
>Lastly, in our drafts any time we mention UE Identification, we put a
>reference to your drafts. We have created a lot of advertisement, PR
>for you. Is it fair to say that this effort went unnoticed? Why should
>we continue to do so if there is no appreciation from your side? And
>we won't.

Med: No comment. 

>
>Regards,
>
>Behcet
>
>On Tue, Oct 16, 2012 at 2:04 AM,  <[e-mail]> wrote:
>> Hi Behcet, Dirk, all,
>>
>> I still think it is too early to ask for a BoF for this 
>work; mainly for the same reasons I listed here:
>> http://www.ietf.org/mail-archive/web/fmc/current/msg00088.html
>>
>> My colleagues and myself, contributed with technical input 
>to be challenged by this list. These inputs were recorded in 
>http://tools.ietf.org/html/draft-schott-fmc-requirements-03. 
>As clearly indicated in that document, it is not voicing for 
>creating a WG but its purpose is to be a place to analyze 
>points which are claimed to be issues.
>>
>> After internal discussion with Sophie, David, Fred and 
>Pierrick, we concluded the main concern is the UE 
>identification. Nevertheless, this issue is not specific to 
>what is called here "fmc" but it is encountered in various use 
>cases. For this reason, we edited this document:
>>  
>http://www.ietf.org/id/draft-boucadair-intarea-host-identifier-
>scenarios-00.txt.
>>
>> We hope that document helped to provide a big picture view 
>of the UE Identification issue.
>>
>> Cheers,
>> Med
>>
>>>-----Message d'origine-----
>>>De : [e-mail] [mailto:[e-mail]] De la
>>>part de Behcet Sarikaya
>>>Envoyé : vendredi 5 octobre 2012 00:29
>>>À : [e-mail]
>>>Objet : [fmc] BoF session in Atlanta
>>>
>>>Hello folks,
>>>
>>>FMC BoF has been scheduled for:
>>>
>>>Tuesday 1700-1830      Afternoon Session III in Salon E
>>>
>>>Regards,
>>>
>>>Dirk & Behcet
>>>_______________________________________________
>>>fmc mailing list
>>>[e-mail]
>>>https://www.ietf.org/mailman/listinfo/fmc
>>>
>