I understand the sentiment. Neither opes nor lemonade are perfect fits for the
current proposals. However, I would offer that the proper home for media
adaptation is either opes or lemonade.
My rationale is simple. Sipping is a transport area work group. The experts
in the group are transport experts. The AD's are transport AD's. Media
transformation is an application. Opes and lemonade are application work
groups. The experts in the groups are application experts. The AD's are
application AD's.
I agree that there may be refinement or conventions that will be needed in SIP
to support real-time media transformation. The correct place for that to
happen is sipping. However, the right people with expertise in media
transformation are in the opes and lemonade groups.
From a charter point of view, media transformation (IMHO) is way out of scope
for sipping. We've heard from Abbie that it is sort of out of scope for opes,
as opes is focusing on HTTP. On the other hand, the mechanism being proposed
is rather close to the lemonade work.
I think this work fits into lemonade charter items 1 (retrieval protocols) and
5 (translation services). We can refine the language to explicitly contain the
real-time adaptation work items, if that would make everyone happy.
--
- Eric
SIPPING is for SIP. OPES and LEMONADE are specifically for media
transformation. Said differently, we have transport experts in SIPPING and
application experts in OPES and LEMONADE.
-----Original Message-----
From: Rohan Mahy [mailto:rohan(_at_)cisco(_dot_)com]
Sent: Wed 11/13/2002 9:00 PM
To: Jose(_dot_)Costa-Requena(_at_)nokia(_dot_)com
Cc: bindignavile(_dot_)srinivas(_at_)nokia(_dot_)com; Eric Burger;
Stephane(_dot_)Coulombe(_at_)nokia(_dot_)com;
jon(_dot_)peterson(_at_)neustar(_dot_)biz;
Markus(_dot_)Isomaki(_at_)nokia(_dot_)com;
dean(_dot_)willis(_at_)softarmor(_dot_)com; drage(_at_)lucent(_dot_)com;
Gonzalo(_dot_)Camarillo(_at_)lmf(_dot_)ericsson(_dot_)se;
sipping(_at_)ietf(_dot_)org; Pekka(_dot_)Pessi(_at_)nokia(_dot_)com;
simple(_at_)mailman(_dot_)dynamicsoft(_dot_)com;
ietf-openproxy(_at_)imc(_dot_)org; UM list
Subject: Re: [Sipping] RE: [Simple] Multimedia message
adaptationInternet-Drafts
hello,
please cease an desist cross posting when you reply. sipping seems
like the default wg, so please send your comments only to
sipping(_at_)ietf(_dot_)org(_dot_)
thanks,
-rohan
On Wednesday, November 13, 2002, at 07:35 AM,
Jose(_dot_)Costa-Requena(_at_)nokia(_dot_)com wrote:
> Hi,
>
> According to LEMONADE requirements, it is considering mainly messaging
> systems and I agree that this proposal could fit into that context at
> some extent. Nevertheless, the actual proposal deals with content
> adaptation based on UA capabilities registered with SIP. Thus, I
> consider that it is within SIPPING scope, as well.
> Comments?
> BR
> Jose
>
> -----Original Message-----
> From: Srinivas Bindignavile (NRC/Boston)
> Subject: RE: [Sipping] RE: [Simple] Multimedia message
> adaptationInternet-Drafts
>
>
> Hi,
>
> As Eric has indicated, the OPES WG is considering transcoding issues!
> However, presently, rather than being protocol-agnostic, it is being
> designed for HTTP and RTP only. SIP is not being considered here.
>
> -Srini
>
>> -----Original Message-----
>> From: ext Eric Burger [mailto:eburger(_at_)snowshore(_dot_)com]
>> Subject: RE: [Sipping] RE: [Simple] Multimedia message
>> adaptationInternet-Drafts
>>
>>
>>
>> There are already TWO work groups that are considering
>> EXACTLY these transcoding requirements.
>>
>> They are OPES and LEMONADE.
>>
>> I would offer that discussion of these capabilities happen in
>> those groups. If SIP is the appropriate mechanism, then
>> those groups will submit the appropriate drafts to SIPPING,
>> outlining the requirements.
>>
>>> -----Original Message----- From:
Jose(_dot_)Costa-Requena(_at_)nokia(_dot_)com
>> [snip]
>>> Nevertheless, "content adaptation" I-D has a wider scope
>>> since it is considering any content-type and it is taking
>>> into account the terminal/user preferences. So I would say
>>> that it fits into SIPPING WG while the filtering I-D is
>>> mainly dealing with presence and I think it should be handled
>>> at SIMPLE WG.
>>> BR
>>> Jose
>>>
>>> -----Original Message----- From: Coulombe Stephane (NRC/Dallas)
>>>> At a high level, these drafts also argue that capability
>>>> negotiation should be administered by intermediaries rather
>>> than through an
>>>> end-to-end process; this approach may attract some similar
>>> controversy.
>>>
>>> Proposed capability negotiation can be used both ways (end-to-end or
>>> administered by intermediaries).
>>> 1) end-to-end: Someone who wants to send an Instant Message
>>> to another user
>>> can send an OPTION query to learn about its terminal
>>> capabilities and
>>> then create a message within its capabilities.
>>>
>>> I guess this is not controversial. However how
>>> realistic and usable is it in practice?
>>> When composing a message, would a user really want to
>>> take into consideration
>>> the image formats to use, message size limitation, etc?
>>>
>>> For instance, you want to send a PNG image to a friend
>>> and his terminal only supports
>>> GIF format. What are you supposed to do? Find an image
>>> conversion tool to convert to GIF?
>>> This is annoying if you are using a PC, imagine with a
>>> mobile phone or handheld?
>>>
>>> For usability reasons, the user wants to send a message
>>> without caring "too much" about
>>> what the other end is supporting.
>>>
>>> 2)administered by intermediaries: this is discussed in detail
>>> in one of the drafts.
>>>
>>> Performing adaptation in the network is controversial
>>> but this is the only way to support
>>> interoperability and good user experience.
>>>
>>>> the applicability and impact of this mechanism is larger
>>> than the problem of
>>>> instant messaging and presence. While clearly, from the
>>> framework, instant
>>>> messaging and presence cases are driving this work, it is
>>> applicable to the
>>>> general use of SIP events (messaging, I think, is something
>>> of a corner case).
>>>
>>> Yes, applicability and impact is larger than IM and presence.
>>> It applies to many other
>>> applications including the case of audio/video conferencing
>>> (for instance when there is
>>> no common audio or video codec between two ends).
>>>
>>> The drafts use the "corner case" of SIP IM for a few reasons:
>>> 1) In SIP IM, there is no concept of capability negotiation
>>> (unlike the case of sessions using SDP).
>>> A user sends a message without knowing anything about
>>> the recipient's terminal capabilities.
>>> 2) In SIP IM, it easier to argue that there will be
>>> interoperability problems because of the variety of content
>>> types that could be sent (in audio/video session codecs are
>>> typically more agreed on). Right now text is mostly used but
>>> richer content will soon be used as is the case in Multimedia
>>> Messaging Service (MMS). By the way, message adaptation is a
>>> serious issue in MMS because of fast product capability
>>> evolution. It's hard to keep interoperability while not
>>> restricting new phones to send just "low-end" content.
>>> 3) It is easier to explain the problem and propose a solution
>>> with a smaller well-defined problem.
>>>
>>> Once we agree that SIP message adaptation is required, the
>>> requirements and solutions should be established from global
>>> perspective; not just SIP IM. For that reason, SIPPING may be
>>> the most appropriate place to initiate this activity.
>>>
>>> Stephane
>>>
>>> -----Original Message-----
>>> From: ext Peterson, Jon
[mailto:jon(_dot_)peterson(_at_)neustar(_dot_)biz]
>>> Sent: Friday, November 08, 2002 6:58 AM
>>> To: Isomaki Markus (NRC/Helsinki);
dean(_dot_)willis(_at_)softarmor(_dot_)com;
>>> drage(_at_)lucent(_dot_)com; rohan(_at_)cisco(_dot_)com;
Gonzalo(_dot_)Camarillo(_at_)lmf(_dot_)ericsson(_dot_)se
>>> Cc: sipping(_at_)ietf(_dot_)org;
simple(_at_)mailman(_dot_)dynamicsoft(_dot_)com
>>> Subject: RE: [Sipping] RE: [Simple] Multimedia message
>>> adaptationInternet-Drafts
>>>
>>>
>>>
>>> It seems to me that these filtering drafts concern the
>>> modification of MIME
>>> bodies in SIP messages by intermediaries. This is not exactly an
>>> uncontroversial topic in SIP circles, and therefore I don't
>>> think it is a
>>> foregone conclusion that this is work that some SIP-related
>> WG should
>>> charter. At a high level, these drafts also argue that capability
>>> negotiation should be administered by intermediaries rather
>>> than through an
>>> end-to-end process; this approach may attract some similar
>>> controversy.
>>>
>>> Provided that this is work the community would like to pursue, the
>>> applicability and impact of this mechanism is larger than the
>>> problem of
>>> instant messaging and presence. While clearly, from the
>>> framework, instant
>>> messaging and presence cases are driving this work, it is
>>> applicable to the
>>> general use of SIP events (messaging, I think, is something
>>> of a corner
>>> case). While SIMPLE could certainly spend some time refining
>>> the framework
>>> and requirements related to IM & presence, I imagine that at
>>> a mechanism
>>> stage this work would need to take place in SIPPING.
>>>
>>> Jon Peterson
>>> NeuStar, Inc.
>>>
>>>> -----Original Message-----
>>>> From: Markus(_dot_)Isomaki(_at_)nokia(_dot_)com
[mailto:Markus(_dot_)Isomaki(_at_)nokia(_dot_)com]
>>>> Sent: Friday, November 08, 2002 3:47 AM
>>>> To: dean(_dot_)willis(_at_)softarmor(_dot_)com;
drage(_at_)lucent(_dot_)com; rohan(_at_)cisco(_dot_)com;
>>>> Gonzalo(_dot_)Camarillo(_at_)lmf(_dot_)ericsson(_dot_)se
>>>> Cc: sipping(_at_)ietf(_dot_)org;
simple(_at_)mailman(_dot_)dynamicsoft(_dot_)com
>>>> Subject: RE: [Sipping] RE: [Simple] Multimedia message
>>>> adaptationInternet-Drafts
>>>>
>>>>
>>>> Hi,
>>>>
>>>> Actually this thread is about two separate things:
>>>> - Event filtering
>>>> - Multimedia message adaptation
>>>>
>>>> Neither of them appears currently on any sippish WG charter
>>>> currently.
>>>>
>>>> Event filtering has been discussed several times and it is
>>>> even mentioned in (but out of scope of) SIP Events RFC. My
>>>> impression has been that people think that it is needed, but
>>>> there has been debate about scope and feasibility. I hope the
>>>> requirements draft will help in that discussion. My own
>>>> opinion is that what is concretely needed in short term is
>>>> some simple filtering definitions for Presence event package.
>>>> More wide-scoped and complex things could be worked upon as
>>>> the understanding accumulates.
>>>>
>>>> Multimedia message adaptation hasn't been yet discussed much.
>>>> I think it is in general a desirable feature, especially for
>>>> relatively small and dumb terminals, which are not easily
>>>> upgradable and may not understand all media formats.
>>>>
>>>> So I propose the WG chairs think where these items would be
>>>> appropriate, and if there is enough interest for them, let's
>>>> put them on the charters.
>>>>
>>>> Markus
>>>>
>>>>> -----Original Message-----
>>>>> From: ext Dean Willis
[mailto:dean(_dot_)willis(_at_)softarmor(_dot_)com]
>>>>> Sent: 08 November, 2002 5:11
>>>>> To: Drage, Keith (Keith)
>>>>> Cc: sipping(_at_)ietf(_dot_)org;
simple(_at_)mailman(_dot_)dynamicsoft(_dot_)com
>>>>> Subject: Re: [Sipping] RE: [Simple] Multimedia message
>>>>> adaptationInternet-Drafts
>>>>>
>>>>>
>>>>> Well, I'd like to hear opinions from the participants here . . .
>>>>>
>>>>> Clearly they aren't explicitly on the charter for either
>>>>> group. Do we as
>>>>> yet have a consensus that we need to work on these
>>>> problems? If so, we
>>>>> can consider WHERE to work on them. I suspect SIPPING is
>>> closer to a
>>>>> matching scope than is SIMPLE, but the relevant ADs may have
>>>>> suggestions
>>>>> to make there as well.
>>>>>
>>>>> --
>>>>> Dean
>>>>>
>>>>> On Wed, 2002-11-06 at 12:51, Drage, Keith (Keith) wrote:
>>>>>> I am getting a bit confused as to which group should be
>>>>> discussing these filtering issues.
>>>>>>
>>>>>> Could we have a statement from the WG chairs of SIPPING or
>>>>> SIMPLE as to whether this, and the moran drafts, are part of
>>>>> the scope of SIPPING or SIMPLE.
>>>>>>
>>>>>> And before you say these are both author drafts, I think we
>>>>> do need to charter one of the WGs to do some work in this
>>>>> area - I am just not sure of the exact scope yet.
>>>>>>
>>>>>> Keith
>>>>>>
>>>>>> Keith Drage
>>>>>> Lucent Technologies
>>>>>> Tel: +44 1793 776249
>>>>>> Email: drage(_at_)lucent(_dot_)com
>>>>>>
>>>>>>> -----Original Message-----
>>>>>>> From: Pekka Pessi
[mailto:Pekka(_dot_)Pessi(_at_)nokia(_dot_)com]
>>>>>>> Sent: 06 November 2002 18:24
>>>>>>> To: sipping(_at_)ietf(_dot_)org;
simple(_at_)mailman(_dot_)dynamicsoft(_dot_)com
>>>>>>> Subject: [Simple] Multimedia message adaptation
>>> Internet-Drafts
>>>>>>>
>>>>>>>
>>>>>>> While these drafts concern event filtering,
>>>> too, the subject was
>>>>>>> a bit misleading because I lazily just followed
>>>> up Tim's e-mail.
>>>>>>>
>>>>>>> Pekka
>>>>>>>
>>>>>>>
>> http://www.ietf.org/internet-drafts/draft-coulombe-message-ada
>>>>>>> ptation-framework-00.txt
>>>>>>>
>>>>>>>
>> http://www.ietf.org/internet-drafts/draft-coulombe-message-ada
>>>>>>> ptation-requirements-00.txt
>>>>>>>
>>>>>>> Pekka Pessi <Pekka(_dot_)Pessi(_at_)nokia(_dot_)com> writes:
>>>>>>>> We have submitted two drafts regarding multimedia message
>>>>>>>> adaptation. A multimedia message is typically a message
>>>>>>>> containing images, audio or video clips and their
>>> presentation
>>>>>>>> information, e.g., smil. Also, even XML-formatted text may
>>>>>>>> require adaptation in some cases.
>>>>>>>
>>>>>>>> Our goal is to have a framework using SIP, HTTP
>> and MIME that
>>>>>>>> allows a person sending multimedia message to adapt
>>> the message
>>>>>>>> contents suitable to all the recipients. In some cases the
>>>>>>>> adaptation can be done by the sending terminal, but
>>> we also see
>>>>>>>> that an adaptation service would be very useful in
>>> many cases.
>>>>>>>> Such an adaptation mechanism is used by MMS service
>>> provided by
>>>>>>>> cellular networks nowadays.
>>>>>>>
>>>>>>>> The message adaptation work concerns both SIPPING
>> and SIMPLE,
>>>>>>>> the requirements I-D lists use cases and requirements for
>>>>>>>> multimedia messaging and message adaptation
>> solutions and the
>>>>>>>> framework I-D tries to explore possible solutions.
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> simple mailing list
>>>>>>> simple(_at_)mailman(_dot_)dynamicsoft(_dot_)com
>>>>>>> http://mailman.dynamicsoft.com/mailman/listinfo/simple
>>>>>>>
>>>>>> _______________________________________________
>>>>>> Sipping mailing list
>>>> https://www1.ietf.org/mailman/listinfo/sipping
>>>>>> This list is for NEW development of the application of SIP
>>>>>> Use sip-implementors(_at_)cs(_dot_)columbia(_dot_)edu for
questions on
>>> current sip
>>>>>> Use sip(_at_)ietf(_dot_)org for new developments of core SIP
>>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> simple mailing list
>>>>> simple(_at_)mailman(_dot_)dynamicsoft(_dot_)com
>>>>> http://mailman.dynamicsoft.com/mailman/listinfo/simple
>>>>>
>>>> _______________________________________________
>>>> simple mailing list
>>>> simple(_at_)mailman(_dot_)dynamicsoft(_dot_)com
>>>> http://mailman.dynamicsoft.com/mailman/listinfo/simple
>>>>
>>> _______________________________________________
>>> Sipping mailing list https://www1.ietf.org/mailman/listinfo/sipping
>>> This list is for NEW development of the application of SIP
>>> Use sip-implementors(_at_)cs(_dot_)columbia(_dot_)edu for questions
on current sip
>>> Use sip(_at_)ietf(_dot_)org for new developments of core SIP
>>> _______________________________________________
>>> simple mailing list
>>> simple(_at_)mailman(_dot_)dynamicsoft(_dot_)com
>>> http://mailman.dynamicsoft.com/mailman/listinfo/simple
>>> _______________________________________________
>>> Sipping mailing list https://www1.ietf.org/mailman/listinfo/sipping
>>> This list is for NEW development of the application of SIP
>>> Use sip-implementors(_at_)cs(_dot_)columbia(_dot_)edu for questions
on current sip
>>> Use sip(_at_)ietf(_dot_)org for new developments of core SIP
>>>
>> _______________________________________________
>> simple mailing list
>> simple(_at_)mailman(_dot_)dynamicsoft(_dot_)com
>> http://mailman.dynamicsoft.com/mailman/listinfo/simple
>>
> _______________________________________________
> Sipping mailing list https://www1.ietf.org/mailman/listinfo/sipping
> This list is for NEW development of the application of SIP
> Use sip-implementors(_at_)cs(_dot_)columbia(_dot_)edu for questions
on current sip
> Use sip(_at_)ietf(_dot_)org for new developments of core SIP
_______________________________________________
Sipping mailing list https://www1.ietf.org/mailman/listinfo/sipping
This list is for NEW development of the application of SIP
Use sip-implementors(_at_)cs(_dot_)columbia(_dot_)edu for questions on
current sip
Use sip(_at_)ietf(_dot_)org for new developments of core SIP