Re: [martini] Call for Consensus: Support for Public GRUUs

[Paul Kyzivat <[e-mail]>]

Richard Shockey wrote:
> -1 
> 
> This statement disturbs me 
> 
> "unless GRUU is mandated in GIN (or any solution to the MARTINI problem, for
> that matter), the chances of GRUU deployment are vanishingly small"
> 
> I don't understand this argument .. so GRUU's have not deployed for well
> known reasons so we want to shove them down the throats of implementers and
> use MARTINI as the mechanism? 
> 
> What in the MARTINI architecture has created this dilema...only the privacy
> issue? 

IMO this is called "learn from you mistakes".
We have a whole hierarchy of mistakes here:

- why hasn't the requirement in 3261 that Contact URIs must be globally
   routable been supported? (This was the reason that public GRUU was
   introduced.)

- why hasn't GRUU been widely implemented?

- if alternatives to GRUU have been deployed, why has none of them
   been proposed as a standard?

If somebody has a different solution for anonymity of calls from the 
sorts of devices that will be using GIN, that can be included/referenced 
in GIN, then I think we should consider that. But I have seen nothing 
other than "that problem will be solved elsewhere".

	Thanks,
	Paul

> -----Original Message-----
> From: [e-mail] [mailto:[e-mail]] On Behalf
> Of Paul Kyzivat
> Sent: Monday, August 30, 2010 9:38 AM
> To: Adam Roach
> Cc: [e-mail]; [e-mail]
> Subject: Re: [martini] Call for Consensus: Support for Public GRUUs
> 
> +1
> 
> Adam Roach wrote:
>> I agree with Andrew on this point
>>
>> The key issue in my mind is deployability. With a single registrar, 
>> deployment of GRUUs is a fairly tractable problem: once the registrar 
>> supports GRUUs, then clients can start making use of them. When we start 
>> having more entities involved in the mix -- like the dual-registrar 
>> situation that MARTINI was chartered to address -- then the chances of 
>> deployment decrease multiplicatively.
>>
>> In other words, unless GRUU is mandated in GIN (or any solution to the 
>> MARTINI problem, for that matter), the chances of GRUU deployment are 
>> vanishingly small, in a way that doesn't exist in a single-registrar 
>> system. Because the MARTINI architecture *caused* this problem, I 
>> believe that it is incumbent on us to *fix* this problem.
>>
>> As public GRUU is really quite easy to implement, I don't imagine that 
>> this requirement should be particularly controversial.
>>
>> /a
>>
>> On 8/29/10 09:28, Aug 29, Andrew Allen wrote:
>>> I certainly agree with the compromise we worked out in Masstricht. As 
>>> I stated during the meeting GRUU is basically a bug fix for SIP and 
>>> without it originally intended basic SIP capabilities either break, 
>>> have negative side effects or have significant limitations on other 
>>> capabilities in order to make them work.
>>>
>>> Public GRUU is not complex to implement and with temporary GRUU the 
>>> compromise gives enough wiggle room to allow other alternatives that 
>>> provide anonymity to be used provided they don't break the use of 
>>> Public GRUUs.
>>>
>>> The privacy text is not so much about implementing privacy as to 
>>> ensure that deployments don't break GRUU capabilities in order to 
>>> achieve privacy.
>>>
>>> Especially in enterprise deployments (which is the major MARTINI use 
>>> case) endpoints will likely not have public IP addresses (or will have 
>>> SBCs that modify them). So the need for GRUU support here is even more 
>>> essential than "generic internet SIP".
>>>
>>> So I agree that SSPs supporting MARTINI MUST implement support for 
>>> providing public GRUUs.
>>>
>>> Andrew
>>>  
>>> *From*: Bernard Aboba [mailto:[e-mail]]
>>> *Sent*: Saturday, August 28, 2010 03:44 PM
>>> *To*: [e-mail] <[e-mail]>
>>> *Subject*: [martini] Call for Consensus: Support for Public GRUUs
>>>  
>>> At the IETF 78 MARTINI WG meeting, during the discussion of Ticket 57 
>>> (relating to temporary GRUUs), a suggestion was made that public GRUUs 
>>> be mandatory to implement for SSPs. 
>>>
>>> We will now attempt to determine whether consensus exists within the 
>>> MARTINI WG to make this change to the document.
>>>
>>> Please respond to this email and post your opinion as to whether you 
>>> agree that SSPs supporting MARTINI MUST implement support for public 
>>> GRUUs.
>>>
>>> This consensus call will last until September 12, 2010. 
>>> ---------------------------------------------------------------------
>>> This transmission (including any attachments) may contain confidential 
>>> information, privileged material (including material protected by the 
>>> solicitor-client or other applicable privileges), or constitute 
>>> non-public information. Any use of this information by anyone other 
>>> than the intended recipient is prohibited. If you have received this 
>>> transmission in error, please immediately reply to the sender and 
>>> delete this information from your system. Use, dissemination, 
>>> distribution, or reproduction of this transmission by unintended 
>>> recipients is not authorized and may be unlawful.
>>>
>>>
>>> _______________________________________________
>>> martini mailing list
>>> [e-mail]
>>> https://www.ietf.org/mailman/listinfo/martini
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> martini mailing list
>> [e-mail]
>> https://www.ietf.org/mailman/listinfo/martini
> _______________________________________________
> martini mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/martini
> 
>