Re: [Ecrit] draft-ietf-ecrit-phonebcp-05.txt: Identifying an emergency call

Hi Brian, 

>Hannes
>
>If it's not possible in most circumstances to have proxies 
>recognize emergency dial strings, then it's not possible to 
>make it a requirement that they do so.
>
>The fact is that the endpoint must do it.  That is the actual 
>requirement.

Quoting the document: 

   ED-3 Endpoints SHOULD recognize dial strings of emergency calls.  If
   the service provider always knows the location of the device, then
   the service provider could recognize them.

>The proxy text is a backstop; it's not required, it's not 
>usually even possible, but it's nice to have.  
>If anything, you could be arguing we should delete the proxy 
>requirement, or at least reword it to cover the case of a 
>guarantee that the visited network handles emergency calls. In 
>fact, since most emergency calls are made from visited 
>networks where the local emergency dial strings are the same 
>as home dial strings, it will actually work most of the time 
>in most networks.
>That's my argument for keeping it.

Henning already made a suggestion for text and I will take a look at
that one to see how it works out. 

Ciao
Hannes

>
>Ergo, SHOULD
>
>Brian
>
>-----Original Message-----
>From: Tschofenig, Hannes (NSN - FI/Espoo) 
>[mailto:hannes.tschofenig@nsn.com]
>
>Sent: Tuesday, October 14, 2008 8:58 AM
>To: ext Brian Rosen; ECRIT
>Subject: RE: [Ecrit] draft-ietf-ecrit-phonebcp-05.txt: 
>Identifying an emergency call
>
>Hi Brian, 
>
>I know this challenge. 
>
>Still, you have to put a must in there to make sure that the 
>call is actually recognized. 
>
>The architecture is very much build on the idea that either 
>the end host or the VSP (directly or indirectly) get location 
>from somewhere. 
>As we learned at the previous emergency services workshops 
>there are some sort of hacks one can do to bypass this need 
>but they only work when there is a relationship between the 
>VSP and the ISP. Currently, these things are being setup in 
>individual countries and as such the question of recognizing 
>emergency service numbers is essentially solved.
>
>
>Ciao
>Hannes
>
>>-----Original Message-----
>>From: ext Brian Rosen [mailto:br@brianrosen.net]
>>Sent: 14 October, 2008 15:51
>>To: Tschofenig, Hannes (NSN - FI/Espoo); 'ECRIT'
>>Subject: RE: [Ecrit] draft-ietf-ecrit-phonebcp-05.txt: 
>>Identifying an emergency call
>>
>>The problem is that proxies don't have location, and therefore can't 
>>recognize them in most cases.  Mobile systems with IMS, where the 
>>visited network handles emergency calls would be an exception, but a 
>>wireline IMS system which allowed nomading would not be able to 
>>recognize local emergency dialstrings.
>>That is why SHOULD is appropriate.
>>
>>Brian
>>
>>-----Original Message-----
>>From: ecrit-bounces@ietf.org [mailto:ecrit-bounces@ietf.org] 
>On Behalf 
>>Of Tschofenig, Hannes (NSN - FI/Espoo)
>>Sent: Tuesday, October 14, 2008 7:49 AM
>>To: ECRIT
>>Subject: [Ecrit] draft-ietf-ecrit-phonebcp-05.txt: Identifying an 
>>emergency call
>>
>>
>>   ED-3 Endpoints SHOULD recognize dial strings of emergency 
>calls.  If
>>   the service provider always knows the location of the device, then
>>   the service provider could recognize them.
>>
>>   SP-2 Proxy servers SHOULD recognize emergency dial strings if for
>>   some reason the endpoint does not recognize them.  This cannot be
>>   relied upon by the device if the service provider cannot always
>>   determine the location of the device.
>>
>>
>>Looking at these 2 SHOULDs I can imagine cases where neither one of 
>>them recognizes them.
>>
>>Wouldn't it be better to have a MUST with the 2nd requirement? 
>>I would delete the 2nd sentence since in that case it would be very 
>>difficult to route the call and to find the right PSAP as well.
>>
>>Comparing with the following requirement I recognize a mismatch: 
>>
>>   ED-19/INT-10/SP-13 Where proxies provide location on behalf of
>>   endpoints, the service provider MUST ensure that either the end
>>   device is provided with the local dial strings for its current
>>   location (where the end device recognizes dial strings), or the
>>   service provider proxy MUST detect the appropriate local dial 
>>strings
>>   at the time of the call. 
>>
>>
>>
>>   ED-5/SP-4 Local dial strings MUST be recognized.
>>
>>"MUST" be recognized by whom? See requirements above. 
>>
>>
>>   ED-6/SP-5 Devices MUST be able to be configured with the home 
>>country
>>   from which the home dial string(s) can be determined.
>>
>>
>>Shouldn't this read like
>>"
>>Devices MUST be have mechanisms to learn the user's home emergency 
>>numbers.
>>"
>>
>>
>>   ED-9 Endpoints SHOULD be able to have home dial strings provisioned
>>   by configuration.
>>
>>I guess ED-9 should say something like "Endpoints SHOULD be able to 
>>have home dial strings dynamically provisioned."
>>To refer to the case that manual configuration may also be used in 
>>order to meet the previous requirements.
>>
>>
>>   ED-11/SP-9 All emergency services specified in [RFC5031] MUST be
>>   recognized.
>>
>>Should say something like: 
>>"
>>All emergency services URNs specified in .... MUST be recognized. 
>>"
>>
>>_______________________________________________
>>Ecrit mailing list
>>Ecrit@ietf.org
>>https://www.ietf.org/mailman/listinfo/ecrit
>>
>>
>
>
_______________________________________________
Ecrit mailing list
Ecrit@ietf.org
https://www.ietf.org/mailman/listinfo/ecrit