Re: [dispatch] Request DISPATCH of RUM

Gunnar Hellström <gunnar.hellstrom@omnitor.se> Mon, 04 February 2019 21:32 UTC

Return-Path: <gunnar.hellstrom@omnitor.se>
X-Original-To: dispatch@ietfa.amsl.com
Delivered-To: dispatch@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B1BF012870E for <dispatch@ietfa.amsl.com>; Mon, 4 Feb 2019 13:32:20 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.901
X-Spam-Level:
X-Spam-Status: No, score=-1.901 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Him5g-xl0peZ for <dispatch@ietfa.amsl.com>; Mon, 4 Feb 2019 13:32:16 -0800 (PST)
Received: from bin-mail-out-06.binero.net (bin-mail-out-06.binero.net [195.74.38.229]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id C6DEF124B0C for <dispatch@ietf.org>; Mon, 4 Feb 2019 13:32:15 -0800 (PST)
X-Halon-ID: 53362001-28c4-11e9-b5ae-0050569116f7
Authorized-sender: gunnar.hellstrom@omnitor.se
Received: from [192.168.2.136] (unknown [83.209.227.185]) by bin-vsp-out-03.atm.binero.net (Halon) with ESMTPSA id 53362001-28c4-11e9-b5ae-0050569116f7; Mon, 04 Feb 2019 22:32:09 +0100 (CET)
To: Brian Rosen <br@brianrosen.net>, Paul Kyzivat <pkyzivat@alum.mit.edu>
Cc: dispatch@ietf.org
References: <q06km0aj8hlr1vn4t5uhe2q7.1549308129655@email.android.com> <789bfa2d-8f8b-55de-5b79-30646bdfafc8@alum.mit.edu> <282ADC66-B79E-4B55-BB6F-F4A95DC7826D@brianrosen.net>
From: =?UTF-8?Q?Gunnar_Hellstr=c3=b6m?= <gunnar.hellstrom@omnitor.se>
Message-ID: <93528f83-43ab-a81e-d89b-c5ed6e33d6f1@omnitor.se>
Date: Mon, 4 Feb 2019 22:32:08 +0100
User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:60.0) Gecko/20100101 Thunderbird/60.5.0
MIME-Version: 1.0
In-Reply-To: <282ADC66-B79E-4B55-BB6F-F4A95DC7826D@brianrosen.net>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 8bit
Content-Language: en-US
Archived-At: <https://mailarchive.ietf.org/arch/msg/dispatch/ri8AquIqjJxSVHA_k0Pvb5r2Nvw>
Subject: Re: [dispatch] Request DISPATCH of RUM
X-BeenThere: dispatch@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: DISPATCH Working Group Mail List <dispatch.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/dispatch>, <mailto:dispatch-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/dispatch/>
List-Post: <mailto:dispatch@ietf.org>
List-Help: <mailto:dispatch-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/dispatch>, <mailto:dispatch-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 04 Feb 2019 21:32:21 -0000

Den 2019-02-04 kl. 21:35, skrev Brian Rosen:

> Yeah, you and I have talked about that, and I think it’s an excellent idea.  Since we can mark calls with language, including various forms of sign language, we could use the interpreter as a transcoder to transcode from, say, video ASL to audio English.

<GH>Yes, I think that is the way to go for future relay services.  It is 
included in a couple of ETSI standards.

The charter might allow it. It mentions the possibility to have a 
provider without an interpreter of its own. That provider can invoke an 
interpreter provider through the provider-to-provider interface, through 
the "dial-around" call case. The provider serving the RUM can also make 
the voice call and link the three legs together.

The model with separation of user handling providers and interpreter 
handling providers is used in a couple of European countries and need to 
be supported by the resulting specification.

I can understand if you did not intend to include the model with the 
user handling provider also making the voice call. But I remember us 
discussing security and authority concerns with the current model 
requiring providers to make calls on behalf of users they have no 
relation with. Maybe it is time to reconsider the model and create a 
clean one.

>
> That isn’t in scope for this work though, which is strictly a regular SIP device which would register with a provider.
>
> I’d like to work on the idea you are proposing; perhaps after we get this work done.
>
> Brian
>
>> On Feb 4, 2019, at 3:05 PM, Paul Kyzivat <pkyzivat@alum.mit.edu> wrote:
>>
>> On 2/4/19 2:32 PM, Gunnar Hellström wrote:
>>> Thanks, I think all your answers are good.
>>> About the WebRTC issue, some clarification in wording seems to be needed.
>>> About the addressing issue, I also think your response is good.
>>> You started that topic with "To my knowledge, there are no Video Interpretation Services that provide interpreters for telephone calls that do not use E.164s as the address."
>>> That is right. My request for support of other addressing forms was rather from the observation that an enormous amount of calls nowadays are set up with other addressing forms than E.164 numbers. With such calls lacking VRS support we are seeing a rapidly decreasing accessibility level in the communication area and a need for improvements.
>> I have been thinking for a long time that it would be good to define a new interface for VRS provision, that only provides the interpretation services, not the telephony origination/termination service. Anybody who wanted to provide audio/video connectivity (regardless of the form of addressing) could then "bridge" in the VRS service when it is needed.
>>
>> 	Thanks,
>> 	Paul
>>
>>> Your conclusion is sufficient. Good.
>>> What wording do you propose for the emergency service support?
>>> Regards
>>> Gunnar
>>> Gunnar Hellström
>>> gunnar.hellstrom@omnitor.se
>>> +46 708 204 288
>>> -------- Originalmeddelande --------
>>> Från: Brian Rosen <br@brianrosen.net>
>>> Datum: 2019-02-04 16:46 (GMT+01:00)
>>> Till: Gunnar Hellström <gunnar.hellstrom@omnitor.se>
>>> Kopia: Paul Kyzivat <pkyzivat@alum.mit.edu>du>, dispatch@ietf.org
>>> Rubrik: Re: [dispatch] Request DISPATCH of RUM
>>> Inline
>>>> On Feb 3, 2019, at 5:12 PM, Gunnar Hellström <gunnar.hellstrom@omnitor.se <mailto:gunnar.hellstrom@omnitor.se>> wrote:
>>>>
>>>> Brian,
>>>>
>>>> A. About technology and WebRTC
>>>> At the moment it seems most important to get the scope clear. And for that, the issue if this is about a WebRTC RUM interface or a native SIP RUM interface or both. Another implementation environment could be 3GPP IMS RUM. Is that also in scope? Is it really realistic to write a RUM spec that fulfilles the goal to allow the users to move a device between providers regardless if they use native SIP or WebRTC (or even IMS )?  ( if that is what is meant by the section about WebRTC).
>>>> Maybe a list of supported call cases is needed to make us understand and accept an intended scope.
>>> I proposed scope that defined a SIP interface where it would be possible to build a WebRTC implementation that ultimately provided the client side of that interface. That means that the WebRTC server created the client side SIP interface towards the provider.
>>> I would like it to be possible to do that, but I don’t think that is the design center.  I do think re-using the media specs from WebRTC is the right set of choices to make for this profile, independent of whether the user has a WebRTC client or a native SIP client.
>>>> B. About Paul's call cases:
>>>> The statement: "For a p2p call with another WebRTC user a gateway won't be
>>>> needed." is in conflict with a statement by the end of the charter saying: "Although the interface between providers also requires standardization to enable multi-provider point-to-point calls, that  interface has already been defined in a SIP Forum document and is thus out of scope for RUM."
>>>> If the users are registered by WebRTC to different providers, and want to set up a p2p call, then the charter says that the providers would use the provider-to-provider interface specified by SIP Forum. But that is a native SIP interface and requires gateways from any WebRTC RUM. So it must be decided, are we specifying for WebRTC, and do we support multi-provider WebRTC calls and do we assume the provider-to-provider interface to be the SIP FORUM spec?
>>> We are specifying a native SIP interface.  It may be possible to interwork some other client to that interface.
>>>> C. Number calling and other forms of calling
>>>> I think the scope regarding addressing should be clarified in the charter. Only one sentence mentions addressing: "The hearing person can also reach D-HOH-SI individuals by in the same manner as calling any other phone number."
>>>> Nowadays a large portion of calls are made with other addressing than phone numbers, and we must prepare to enable VRS use also in such calls. It would be best if the RUM spec is agnostic to what kind of addressing is used. That may be possible, because the number handling and conversion between numbers and SIP URI is handled elsewhere. I think anyway that the scope of addressing variants is worth a paragraph in the charter.
>>> To my knowledge, there are no Video Interpretation Services that provide interpreters for telephone calls that do not use E.164s as the address.   While the actual address of the SIP device is a SIP URI, and we could say that we support that, the interwork for dialing needs to be explicit, so that whenever telephone numbers are used, the device interface uses them consistently.    I’ll propose language that allows SIP URIs for the device address.
>>>> Proposal: Delete the mentioning of number, and include the following paragraph: "Addressing of participants in the calls are supposed to be based on general addressing conventions in SIP. Any conversion needed between this form and other addressing forms (e.g. phone numbers) required for completion of the calls are assumed to take place in other parts of the networks."
>>>>
>>>> D: Do you also want to discuss minor edits in the charter? If so, here are a couple of hints:
>>>>
>>>> D.1. This phrase: "The deaf, hard- of- hearing or speech-impaired person (D-HOH-SI) uses a SIP-based video phone to connect with an interpreter, and the interpreter places a phone call on the PSTN to the hearing person." should be changed. PSTN is fading away. The call might also be automated. Maybe it is sufficient to modify it to:  "The deaf, hard- of- hearing or speech-impaired person (D-HOH-SI) uses a SIP-based video phone to connect with a provider, and the provider conveyes the call to the hearing person and includes an interpreter in the call."
>>>>
>>>> (this wording would also support the implementation when the interpreter actually is involved in placing the leg of the call to the hearing person)
>>> I will adjust the wording
>>>> D.2. There are a number of places where it is obvious that the wording is copied from a  document with changemarks, and both the original and the modification happened to be included in the charter. The first example of this kind is on line 4, where "VRSwhich" has apparently history in a modification from "which" to "VRS" and should be just  "VRS". I can help to sort them out if you want.
>>> Yes, sorry, I didn’t notice that.  I will fix it in the next version
>>>>
>>>> Regards
>>>>
>>>> Gunnar
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> -------- Originalmeddelande --------
>>>> Från: Paul Kyzivat <pkyzivat@alum.mit.edu <mailto:pkyzivat@alum.mit.edu>>
>>>> Datum: 2019-02-03 02:29 (GMT+01:00)
>>>> Till:dispatch@ietf.org <mailto:dispatch@ietf.org>
>>>> Rubrik: Re: [dispatch] Request DISPATCH of RUM
>>>>
>>>> On 2/2/19 6:18 PM, Brian Rosen wrote:
>>>>> Hmmm, yeah, we really need to keep RTT in RTP.
>>>>> We’ll have to use a gateway for that
>>>> In the case of a call with an interpreter the provider can provide the
>>>> gateway, or provide the interpreter with native text over data channel
>>>> support. For a p2p call with another WebRTC user a gateway won't be
>>>> needed. The only special case is a p2p call with one WebRTC endpoint and
>>>> one native SIP endpoint.
>>>>
>>>> Thanks,
>>>> Paul
>>>>
>>>>> Yes emergency calls are in scope.
>>>>>
>>>>> Brian
>>>>>
>>>>> On Sat, Feb 2, 2019 at 5:48 PM Gunnar Hellström
>>>>> <gunnar.hellstrom@omnitor.se <mailto:gunnar.hellstrom@omnitor.se><mailto:gunnar.hellstrom@omnitor.se>> wrote:
>>>>>
>>>>>      So far, in tradiitional SIP based VRS, real-time text has been
>>>>>      implemented with RTP, while in WebRTC it is supposed to use the data
>>>>>      channel. How would you specify that interop without a media gateway?
>>>>>
>>>>>
>>>>>      Another issue: should possibility to interop with emergency services
>>>>>      be mentioned in the charter? I assume that such calls need to pass
>>>>>      through the provider, and can be gatewayed ther, but there is a
>>>>>      desire that all media is conveyed between the emergency service and
>>>>>      the RUM and  there might therefore be a need to consider this
>>>>>      requirement when specifying the RUM interface.
>>>>>
>>>>>
>>>>>      Regards
>>>>>      Gunnar
>>>>>
>>>>>
>>>>>
>>>>>      -------- Originalmeddelande --------
>>>>>      Från: Christer Holmberg <christer.holmberg@ericsson.com <mailto:christer.holmberg@ericsson.com>
>>>>>      <mailto:christer.holmberg@ericsson.com>>
>>>>>      Datum: 2019-02-02 00:10 (GMT+01:00)
>>>>>      Till: Brian Rosen <br@brianrosen.net <mailto:br@brianrosen.net><mailto:br@brianrosen.net>>
>>>>>      Kopia: DISPATCH list <dispatch@ietf.org <mailto:dispatch@ietf.org><mailto:dispatch@ietf.org>>
>>>>>      Rubrik: Re: [dispatch] Request DISPATCH of RUM
>>>>>
>>>>>
>>>>>      Hi,
>>>>>
>>>>>       >Yes, that’s the idea.  I will work on some wording. I don’t want
>>>>>      the charter to have a
>>>>>       >list of such features.
>>>>>
>>>>>      You could say that the profile will mandate all features needed in
>>>>>      order to interoperate with WebRTC without having to use a media
>>>>>      gateway, or something like that.
>>>>>
>>>>>      Regards,
>>>>>
>>>>>      Christer
>>>>>
>>>>>
>>>>>
>>>>>      Brian
>>>>>
>>>>>      On Fri, Feb 1, 2019 at 5:30 PM Christer Holmberg
>>>>>      <christer.holmberg@ericsson.com <mailto:christer.holmberg@ericsson.com>
>>>>>      <mailto:christer.holmberg@ericsson.com>> wrote:
>>>>>
>>>>>          Hi,
>>>>>
>>>>>           >Can you suggest a wording change?
>>>>>
>>>>>          Not at the moment, I first want to understand exactly what the
>>>>>          scope and purpose is.
>>>>>
>>>>>           >It now says "A WebRTC- based RUM could create a SIP interface
>>>>>          (using, e.g., SIP over
>>>>>           > WebSockets) towards the provider that conformed to the
>>>>>          document RUM will produce.  Such >an implementation should be
>>>>>          possible, ideally without requiring a media gateway.”  That
>>>>>           >seems to me to be clear that the wg won’t do any work beyond
>>>>>          making sure it’s possible to >create a WebRTC based RUM without
>>>>>          a media gateway.
>>>>>
>>>>>          If the WG is going to "make sure" that it works without a media
>>>>>          gateway, does that mean that you would also mandate e.g., ICE,
>>>>>          continuous consent, DTLS, and whatever other media level
>>>>>          features might be mandated to support by WebRTC? If so, I think
>>>>>          it needs to be very clear.
>>>>>
>>>>>          Regards,
>>>>>
>>>>>          Christer
>>>>>
>>>>>
>>>>>          Brian
>>>>>
>>>>>
>>>>>
>>>>>>          On Feb 1, 2019, at 4:57 PM, Christer Holmberg
>>>>>>          <christer.holmberg@ericsson.com <mailto:christer.holmberg@ericsson.com>
>>>>>>          <mailto:christer.holmberg@ericsson.com>> wrote:
>>>>>>
>>>>>>
>>>>>>          Hi,
>>>>>>
>>>>>>          >We want to make sure the mechanisms interact reasonably.  We
>>>>>>          suspect this is mostly codec
>>>>>>          >choices, etc.
>>>>>>
>>>>>>          Then you should say that a goal is interoperability with
>>>>>>          WebRTC when it comes to codecs etc.
>>>>>>
>>>>>>          The way I read the text now is that the WG is about writing
>>>>>>          WebRTC SIP clients, which I assume is outside the scope😊
>>>>>>
>>>>>>          Regards,
>>>>>>
>>>>>>          Christer
>>>>>>
>>>>>>
>>>>>>>          On Feb 1, 2019, at 4:11 PM, Christer Holmberg
>>>>>>>          <christer.holmberg@ericsson.com <mailto:christer.holmberg@ericsson.com>
>>>>>>>          <mailto:christer..holmberg@ericsson.com>> wrote:
>>>>>>>
>>>>>>>          Hi,
>>>>>>>
>>>>>>>          If the purpose of the WG is to define a SIP profile, I assume
>>>>>>>          it does not matter if the SIP UAs are implemented using
>>>>>>>          WebRTC or something else, so why does the charter need to
>>>>>>>          talk about WebRTC?
>>>>>>>
>>>>>>>          If you want to look at some of the specific network
>>>>>>>          technologies used by WebRTC, e.g., the data channel, I think
>>>>>>>          should talk about that instead.
>>>>>>>
>>>>>>>          Regards,
>>>>>>>
>>>>>>>          Christer
>>>>>>>
>>>>>>>
>>>>>>>          ------------------------------------------------------------------------
>>>>>>>          *From:*dispatch <dispatch-bounces@ietf.org <mailto:dispatch-bounces@ietf.org>
>>>>>>>          <mailto:dispatch-bounces@ietf.org>> on behalf of Brian Rosen
>>>>>>>          <br@brianrosen.net <mailto:br@brianrosen.net><mailto:br@brianrosen.net>>
>>>>>>>          *Sent:*Friday, February 1, 2019 10:50:53 PM
>>>>>>>          *To:*DISPATCH list
>>>>>>>          *Subject:*[dispatch] Request DISPATCH of RUM
>>>>>>>          I would like dispatch to consider spinning up a mini-work
>>>>>>>          group to create a sip device profile for use with Video Relay
>>>>>>>          Services.
>>>>>>>
>>>>>>>
>>>>>>>          The proposed charter is:
>>>>>>>
>>>>>>>          Relay User Machine (rum) Working Group Proposed Charter
>>>>>>>          ART Area
>>>>>>>
>>>>>>>          Many current instances of Video Relay Service (VRS),
>>>>>>>          (sometimes called Video Interpretation Service.), use the
>>>>>>>          Session Initiation Protocol (SIP) and other IETF multimedia
>>>>>>>          protocols. VRSwhich is used bya service that deaf and hard-
>>>>>>>          of- hearing persons and person with speech impairments use to
>>>>>>>          communicate with hearing persons.  The deaf, hard- of-
>>>>>>>          hearing or speech-impaired person (D-HOH-SI) uses a SIP-
>>>>>>>          based video phone to connect with an interpreter, and the
>>>>>>>          interpreter places a phone call on the PSTN to the hearing
>>>>>>>          person. The hearing person can also reach D-HOH-SI
>>>>>>>          individuals by in the same manner as calling any other phone
>>>>>>>          number.  The D-HOH-SI person uses sign language and possibly
>>>>>>>          real-time text with the interpreter and the interpreter uses
>>>>>>>          spoken language with the hearing person, providing on- line,
>>>>>>>          real- time, two- way communication.  VRS services are often
>>>>>>>          government- supported.  In some countries, private companies
>>>>>>>          provide the interpreters, and compete with one another.
>>>>>>>          Often these companies use proprietary implementations for
>>>>>>>          user devices as a means of vendor lock in.
>>>>>>>
>>>>>>>          Having a standard interface between the end- user device and
>>>>>>>          the VRS provider allows vendors and open-source developers to
>>>>>>>          build devices that work with multiple service providers;
>>>>>>>          devices can also be retained when changing providers.  In
>>>>>>>          this instance, “device” could be a purpose-built videophone
>>>>>>>          or could be downloadable software on a general purpose
>>>>>>>          computing platform or mobile phone.  The SIP protocol is
>>>>>>>          complex enough that some form of profiling is needed to
>>>>>>>          achieve interoperability between devices and providers. To
>>>>>>>          ensure interoperability of the key features of this service,
>>>>>>>          certain aspects (e.g., codecs, media transport, and SIP
>>>>>>>          features) must be specified as mandatory-to-implement for
>>>>>>>          SIP-based VRS devices. These specified features effectively
>>>>>>>          form a profile for SIP and the media it negotiates.
>>>>>>>
>>>>>>>          This working group will produce a single document: a profile
>>>>>>>          of SIP and media features for use with video relay services
>>>>>>>          (which includes video, real time text, and audio), and other
>>>>>>>          similar interpretation services that require multimedia.  It
>>>>>>>          will reference the IETF’s current thinking on multimedia
>>>>>>>          communicationsuch devices, including references to work
>>>>>>>          beyond SIP (e.g., WebRTC and SLIM).  No protocol changes are
>>>>>>>          anticipated by this work.
>>>>>>>
>>>>>>>          While WebRTC could be used to implement a RUM, the group’s
>>>>>>>          work will focusis on the device-to-provider interface.  A
>>>>>>>          WebRTC- based RUM couldwould create a SIP interface (using,
>>>>>>>          e.g., SIP over WebSockets) towards the provider that
>>>>>>>          conformed to the document RUMrum will produce.  Such an
>>>>>>>          implementation should be possible, ideally without requiring
>>>>>>>          a media gateway.
>>>>>>>
>>>>>>>          The scope of the work includes mechanisms to provision the
>>>>>>>          user’s device with common features such as speed dial lists,
>>>>>>>          provider to contact, videomail service interface point and
>>>>>>>          similar items.  These features allow users to more easily
>>>>>>>          switch providers temporarily (a feature known as “dial
>>>>>>>          around”) or permanently, while retaining their data.
>>>>>>>
>>>>>>>          Devices used in VRS can be used to place point- to- point
>>>>>>>          calls, i.e., where both communicating parties use sign
>>>>>>>          language.  When used for point-to-point calling where the
>>>>>>>          participants are not served by the same VRS provider, or when
>>>>>>>          one provider provides the originating multimedia transport
>>>>>>>          environment, but another provides the interpreter
>>>>>>>          (“dial-around call”), the call traverses two providers.  Both
>>>>>>>          of these uses impose additional requirements on a RUMrum
>>>>>>>          device and are in scope for this work.
>>>>>>>
>>>>>>>          Although the interface between providers also requires
>>>>>>>          standardization to enable multi-provider point-to-point
>>>>>>>          calls, that  interface has already been defined in a SIP
>>>>>>>          Forum document and is thus out of scope for RUM.
>>>>>>>          _______________________________________________
>>>>>>>          dispatch mailing list
>>>>>>>         dispatch@ietf.org <mailto:dispatch@ietf.org><mailto:dispatch@ietf.org>
>>>>>>>         https://www.ietf.org/mailman/listinfo/dispatch
>>>>>
>>>>> _______________________________________________
>>>>> dispatch mailing list
>>>>> dispatch@ietf.org <mailto:dispatch@ietf.org>
>>>>> https://www.ietf.org/mailman/listinfo/dispatch
>>>>>
>>>> _______________________________________________
>>>> dispatch mailing list
>>>> dispatch@ietf.org <mailto:dispatch@ietf.org>
>>>> https://www.ietf.org/mailman/listinfo/dispatch
>>>> _______________________________________________
>>>> dispatch mailing list
>>>> dispatch@ietf.org <mailto:dispatch@ietf.org>
>>>> https://www.ietf.org/mailman/listinfo/dispatch

-- 
-----------------------------------------
Gunnar Hellström
Omnitor
gunnar.hellstrom@omnitor.se
+46 708 204 288