Re: [Ecrit] 3GPP impact on draft-ecall: usage of legacy modem to transport MSD

Hi Keith,

Especially since the car-crash draft references this draft and uses 
the same URNs (per your request), it might be best to leave the URN 
registration here, and add text such as "The semantics of these URNs 
when used with an IMS-based eCall are defined by 3GPP."

The draft already registers "ecall" as a child of the "test" service.

--Randy

At 4:24 PM +0000 6/8/16, Keith (Nokia - GB) Drage wrote:

>  Going the direction of 3GPP definition means the draft should 
> remove the URNs entirely and leave their definition and 
> registration to 3GPP. The document definiting the semantics should 
> be referenced by the IANA registration.
>
>  In terms of text that we need, my suggestion as a starting point 
> would be something along the lines of:
>
>  "ecall: This subtype of the sos URN represents a resource that is 
> optimsed for handling emergency calls carrying ecall data. This URN 
> represents a resource that supports both the data transfer 
> mechanisms identified in this draft, and also those defined in 3GPP 
> TS 26.267 [ref]."
>
>  You will need to work this definition into both 14.1 and into 
> either clause 6 (area of last paragraph) or clause 7.
>
>  Note that "ecall" is a subtype in its own right, and will need to 
> defined separately to "manual" and "automatic", which are also 
> subtypes in their own right. Theorectically, in terms of what you 
> have defined sos.ecall is equally valid as sos.ecall.manual and 
> sos.ecall.automatic.
>
>  I notice that you will probably also need to register "ecall" as a 
> subtype of "test.sos" in the IANA definitions.
>
>  Regards
>
>  Keith
>
>  -----Original Message-----
>  From: Randall Gellens [mailto:rg+ietf@randy.pensive.org]
>  Sent: 08 June 2016 16:57
>  To: Drage, Keith (Nokia - GB); Christer Holmberg; ecrit@ietf.org
>  Subject: RE: [Ecrit] 3GPP impact on draft-ecall: usage of legacy 
> modem to transport MSD
>
>  Hi Keith,
>
>  It seems to me that 3GPP is the entity that should define the 
> semantics, since they define IMS and IMS emergency calls.
>
>  If you feel that the draft needs to do so, could you send the text 
> you suggest be added to the draft, so we can see it and see if it 
> is in scope for the draft?
>
>  --Randy
>
>  At 12:27 PM +0000 6/8/16, Keith (Nokia - GB) Drage wrote:
>
>>   I have made similar comments to Christer at least a couple of times.
>>
>>   Firstly, the only usage of this draft is in 3GPP networks. Noone has
>>  identified a use case or a deployment for this draft outside 3GPP.
>>  3GPP has special usages of SIP for emergency calls, which essentially
>>  require it to be used via IMS entities. Try and use SIP outside this,
>>  and it will not be an emergency call in the 3GPP network, and the call
>>  will probably fail for roaming users. The IMS standards do specify
>>  what happens in regard to correlation with the CS domain, and the URNs
>>  defined for emergency call usage must take account of it.
>>
>>   The URNs identity the resource to which the call should be routed.
>>
>>   For the existing sos URNs, the interworking with the CS domain has no
>>  special capabilities and is handled within IMS. The media bearers used
>>  for IMS based emergency calls are a superset of those used for CS
>>  domain emergency call, and therefore there is no disparity in the URN
>>  usage.
>>
>>   In 3GPP ecall, that resource must be a resource capable of handling
>>  both the data transfer mechanism specifically designed for IMS, but
>>  also the existing data transfer mechanism using a modem based
>>  mechanism. There is no interworking point defined for converting one
>>  mechnanism to the other, and the PSAP is expected to handle both (as
>>  is the calling UA). Therefore the URNs must identity a resource that
>>  handles both. If transmission fails using the mechanism defined by
>>  this draft (presumably because of
>>  interworking) then the UA and PSAP will drop back to trying to use the
>   > existing modem based approach.
>>
>>   Therefore the draft that defined the URN must define these semantics
>>  for the resource addressed.
>>
>>   No one is saying you need to write the rest of the story relating to
>>  CS data transfer, but the URN definition must encompass it.
>>
>>   Regards
>>
>>   Keith
>>
>>   -----Original Message-----
>>   From: Ecrit [mailto:ecrit-bounces@ietf.org] On Behalf Of Randall
>>  Gellens
>>   Sent: 08 June 2016 01:03
>>   To: Christer Holmberg; ecrit@ietf.org
>>   Subject: Re: [Ecrit] 3GPP impact on draft-ecall: usage of legacy
>>  modem to transport MSD
>>
>>   Hi Christer,
>>
>>   The URNs are registered as child elements of the SOS service URN, so
>>  they are in no way limited to INFO.
>>
>>   At 5:27 AM +0000 6/5/16, Christer Holmberg wrote:
>>
>>>    Clarification: the URNs cannot be limited to the INFO mechanism for 
>>>  transporting MSD.
>>>
>>>    Regards,
>>>
>>>    Christer
>>>
>>>    Sent from my Windows Phone
>>>
>>>    From: <mailto:christer.holmberg@ericsson.com>Christer Holmberg
>>>    Sent: 05/06/2016 08:21
>>>    To: <mailto:rg+ietf@randy.pensive.org>Randall Gellens; 
>>>  <mailto:ecrit@ietf.org>ecrit@ietf.org
>>>    Subject: Re: [Ecrit] 3GPP impact on draft-ecall: usage of legacy 
>>>  modem to transport MSD
>>>
>>>    Hi,
>>>
>>>    The draft also defines eCall URNs, and those cannot be limited to SIP.
>>>
>>>    Regards,
>>>
>>>    Christer
>>>
>>>    Sent from my Windows Phone
>>>
>>>    From: <mailto:rg+ietf@randy.pensive.org>Randall Gellens
>>>    Sent: 05/06/2016 02:49
>>>    To: <mailto:christer.holmberg@ericsson.com>Christer Holmberg; 
>>>  <mailto:ecrit@ietf.org>ecrit@ietf.org
>>>    Subject: Re: [Ecrit] 3GPP impact on draft-ecall: usage of legacy 
>>>  modem to transport MSD
>>>
>>>    Hi Christer,
>>>
>>>    I think in-band modem is out of scope of the draft.  The draft is 
>>>  limited to SIP signaling aspects.
>>>
>>>    --Randy
>>>
>>>    At 11:35 AM +0000 6/3/16, Christer Holmberg wrote:
>>>
>>>>     Hi,
>>>>
>>>>     3GPP agreed that, when communicating with a PSTN PSAP, it shall
>>>>  be
>>    >>   possible to transport MSD using a legacy modem on the media 
>> plane. The
>>>>     reason is for not mandating the network to perform 
>>>> "interworking between
>>>>     MSD transported in INVITE/INFO and MSD transported using a legacy
>>>>  modem on
>>>>     the media plane".
>>>>
>>>>     I don't think the ecall draft needs to consider WHEN the MSD will be
>>>>     transported using a legacy modem on the media plane, but I think
>>>>  the draft
>>>>     should describe the possibility, so that PSAPs are aware of it.
>>>>
>>>>     Also, as the same eCall URNs will be used for calls where the MSD is
>>>>     transported using a legacy modem on the media plane, the draft
>>>>  must enable
>>>>     usage of the eCall URNs in setup of an IMS emergency call where MSD is
>>>>     transported using the legacy modem on the media plane.
>>>>
>>>>     Regards,
>>>>
>>>>     Christer
>>>>
>>>>     _______________________________________________
>>>>     Ecrit mailing list
>>>>     Ecrit@ietf.org
>>>>
>>>> 
>>>>  <https://www.ietf.org/mailman/listinfo/ecrit>https://www.ietf.org/ma
>>>>  i
>>>>   lman/listinfo/ecrit
>>>
>>>
>>>    --
>>>    Randall Gellens
>>>    Opinions are personal;    facts are suspect;    I speak for myself only
>>>    -------------- Randomly selected tag: ---------------
>>>>From new transmitters came the old stupidities.   --Bertolt Brecht
>>
>>
>>   --
>>   Randall Gellens
>>   Opinions are personal;    facts are suspect;    I speak for myself only
>>   -------------- Randomly selected tag: --------------- True patriotism
>>  hates injustice in its own land more than anywhere else.
>>       --Clarence Darrow
>>
>>   _______________________________________________
>>   Ecrit mailing list
>>   Ecrit@ietf.org
>>   https://www.ietf.org/mailman/listinfo/ecrit
>
>
>  --
>  Randall Gellens
>  Opinions are personal;    facts are suspect;    I speak for myself only
>  -------------- Randomly selected tag: --------------- 37% of 
> Americans agree that while they would hate being British, they 
> wouldn't mind having a British accent.

-- 
Randall Gellens
Opinions are personal;    facts are suspect;    I speak for myself only
-------------- Randomly selected tag: ---------------
Justice is incidental to law and order.
                      --J.Edgar Hoover