Re: [Slim] Indication of modality alternatives in draft-ietf-slim-negotiating-human-language -Issue #46

[Gunnar Hellström <[e-mail]>]

In order to not create complicated sentences but still having the 
wording match our intentions, I want to change the proposed resolution 
for Issue # 46 Change 1 to:

---Change 1 in 5.2, first paragraph----------------
------old text---------
This document defines two media-level attributes starting with
'hlang' (short for "human interactive language") to negotiate which
human language is selected for use in each interactive media stream.
------------new text--------------------
This document defines two media-level attributes starting with
'hlang' (short for "human interactive language") to negotiate which
human language is selected for *potential* use in each media stream.
-------end of change 1-------

That matches the "if" in paragraph 3, and it is also valid for both the 
offers and answers, while paragraph 3 is only for the answer.
Please accept it, it is of importance for proper understanding of our 
intentions.

/Gunnar

Den 2017-10-14 kl. 00:21, skrev Gunnar Hellström:
> Den 2017-10-13 kl. 20:31, skrev Brian Rosen:
>> Gunnar
>>
>> Protocol documents are for engineers to write software/create 
>> hardware.  They don’t try to control user behavior.  I think in this 
>> case, you are trying to get the document to describe user behavior 
>> and not implementation software/hardware.
>>
>> Although we do sometimes describe how we expect the protocol to be 
>> used by people, that is not normative, and we should be careful to 
>> not proscribe behavior.
> <GH>Our protocol needs to be well defined regardless if the source and 
> sink of language is automata or humans.
> As long as we can read the specification differently it is not well 
> defined.
> When I ask what the result of the negotiation really means, you use to 
> say that it is alternative languages that the users are supposed to 
> select from and use one or more in each direction.
> I agree that that is a good result.
> I think the wording still means that all negotiated languages should 
> be used, and I want to avoid that interpretation.
>
> (There is also use for a result saying that a couple of languages are 
> desired together in the same direction, but it has been said many 
> times that that is not the intention of the current draft, so that 
> requires separate work. )
>
> We are reasonably good now with change 2 implemented. The first 
> section of 5.2, that change 1 aims at, may be seen as introductory and 
> does maybe not need to be completely clarifying, if that requires too 
> complicated wording. But it must not contradict the intention of the 
> protocol.
>
> Gunnar
>>
>> Brian
>>
>>> On Oct 13, 2017, at 2:21 PM, Gunnar Hellström 
>>> <[e-mail] <mailto:[e-mail]>> 
>>> wrote:
>>>
>>> Den 2017-10-13 kl. 16:58, skrev Bernard Aboba:
>>>> Gunnar said:
>>>>
>>>> "to negotiate which human language is selected for possible use in 
>>>> each interactive media stream"
>>>>
>>>> [BA] Given that audio can be muted, video can be turned off, etc. 
>>>> aren't media streams negotiated in SDP always for "possible" use?
>>> <GH>That may be true, but we are not talking about the media flow in 
>>> the streams. We are talking about the use for language. Our draft 
>>> must reflect clearly what the language negotiation result really 
>>> means. To me,  "is selected for use in each interactive media 
>>> stream" sounds as a promise that a negotiated language will actually 
>>> be used. That means that if two media streams end up with negotiated 
>>> languages in the same direction, then both must be provided 
>>> together. According to the discussions in the WG, that is not the 
>>> desired result. The desired result should be that the users can 
>>> select between use of the negotiated languages and usually use just 
>>> one in each direction.  We introduced "selected" some time ago, but 
>>> it did not have the right effect.
>>>
>>> I will try to come up with new wording proposals.
>>>
>>> /Gunnar
>>>
>>>>
>>>> On Fri, Oct 13, 2017 at 6:32 AM, Gunnar 
>>>> Hellström<[e-mail] 
>>>> <mailto:[e-mail]>>wrote:
>>>>
>>>>     Change 2 is fine and solves part of the problem.
>>>>
>>>>     But the current wording at my proposed change 1 still tells me
>>>>     that if I offer English text and English voice, it means that I
>>>>     have selected to use both, and even stronger if an answer
>>>>     contains English text and English voice, then both will be used
>>>>     in the session, exactly as you indicated was the problem with
>>>>     the Lang attribute. We need to get the possibility to select
>>>>     among alternatives clearly into the draft so that not next
>>>>     generation implementers also say that it is too vague about
>>>>     what it means.
>>>>
>>>>     The current wording at change one still says that each
>>>>     interactive stream is used.
>>>>
>>>>     How about:  "to negotiate which
>>>>     human language is selected for*possible*use in each interactive
>>>>     media stream."
>>>>
>>>>     /Gunnar
>>>>
>>>>
>>>>     Den 2017-10-13 kl. 15:13, skrev Randall Gellens:
>>>>>     I think we've addressed the concerns that existed with earlier
>>>>>     versions of the draft.
>>>>>
>>>>>     At 2:57 PM +0200 10/13/17, Gunnar Hellström wrote:
>>>>>
>>>>>>      Den 2017-10-13 kl. 13:51, skrev Randall Gellens:
>>>>>>>      At 12:06 AM +0200 7/29/17, Gunnar Hellström wrote:
>>>>>>>
>>>>>>>>     We have dealt with this topic before, but rereading the
>>>>>>>>     draft indicates to me that we still need some tuning of the
>>>>>>>>     wording so that it is clear that the language indications
>>>>>>>>     for the same direction for different media are alternatives
>>>>>>>>     with no requirements that they need to be provided
>>>>>>>>     together, so that it is allowed to answer with just one
>>>>>>>>     media in each direction having language indication.
>>>>>>>>
>>>>>>>>     Suggested wording changes to make this clear:
>>>>>>>>
>>>>>>>>     ---Change 1 in 5.2, first paragraph----------------
>>>>>>>>     ------old text---------
>>>>>>>>     This document defines two media-level attributes starting with
>>>>>>>>     'hlang' (short for "human interactive language") to
>>>>>>>>     negotiate which
>>>>>>>>     human language is selected for use in each interactive
>>>>>>>>     media stream.
>>>>>>>>     ------------new text--------------------
>>>>>>>>     This document defines two media-level attributes starting with
>>>>>>>>     'hlang' (short for "human interactive language") to
>>>>>>>>     negotiate which
>>>>>>>>     human language is selected for use in each media stream
>>>>>>>>     used for interactive language communication.
>>>>>>>>     -------end of change 1-------
>>>>>>>
>>>>>>>      I don't see how changing "each interactive media stream" to
>>>>>>>     "each media stream used for interactive language
>>>>>>>     communication" improves anything.  The term "interactive"
>>>>>>>     implies human interaction.
>>>>>>      <GH>Yes, but human interaction can be to show things in
>>>>>>     video without being language communication.
>>>>>>      What I am aiming at is to clearly indicate that the language
>>>>>>     indications are alternatives to select from. The wording "use
>>>>>>     in each interactive media stream" sounds to me that you MUST
>>>>>>     use all the agreed languages. That is the same mistake that
>>>>>>     you initially blamed the Lang SDP attribute to mean. We need
>>>>>>     to get away from that interpretation. My wording was intended
>>>>>>     to accomplish that, but it might have been too weak. The key
>>>>>>     word is "used" that is intended to mean that if a media
>>>>>>     stream is selected to be used for language communication then
>>>>>>     the agreed language is the one to be used.
>>>>>>      So, I prefer my wording, or if you can create something even
>>>>>>     more clear that we are talking about alternatives to select from.
>>>>>>>
>>>>>>>>
>>>>>>>>     ----Change 2 in 5.2, third paragraph ------
>>>>>>>>     ----old text------
>>>>>>>>     In an answer, 'hlang-send' is the language the answerer
>>>>>>>>     will send if
>>>>>>>>     using the media for language (which in most cases is one of the
>>>>>>>>     languages in the offer's 'hlang-recv'), and 'hlang-recv' is the
>>>>>>>>     language the answerer expects to receive in the media
>>>>>>>>     (which in most
>>>>>>>>     cases is one of the languages in the offer's 'hlang-send').
>>>>>>>>     -----new text----
>>>>>>>>     In an answer, 'hlang-send' is the language the answerer
>>>>>>>>     will send if
>>>>>>>>     using the media for language (which in most cases is one of the
>>>>>>>>     languages in the offer's 'hlang-recv'), and 'hlang-recv' is the
>>>>>>>>     language the answerer expects to receive in the media if
>>>>>>>>     using the media for language (which in most
>>>>>>>>     cases is one of the languages in the offer's 'hlang-send').
>>>>>>>>     ----end of change 2-------------------------------
>>>>>>>
>>>>>>>      I'm OK adding "if using the media for language" to the
>>>>>>>     second clause.
>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>     /Gunnar
>>>>>>>>
>>>>>>>>     --
>>>>>>>>     -----------------------------------------
>>>>>>>>     Gunnar Hellström
>>>>>>>>     Omnitor
>>>>>>>>     [e-mail]
>>>>>>>>     <mailto:[e-mail]>
>>>>>>>>
>>>>>>>>     _______________________________________________
>>>>>>>>     SLIM mailing list
>>>>>>>>     [e-mail] <mailto:[e-mail]>
>>>>>>>>     https://www.ietf.org/mailman/listinfo/slim
>>>>>>>>     <https://www.ietf.org/mailman/listinfo/slim>
>>>>>>>
>>>>>>>
>>>>>>
>>>>>>      --
>>>>>>      -----------------------------------------
>>>>>>      Gunnar Hellström
>>>>>>      Omnitor
>>>>>>     [e-mail] <mailto:[e-mail]>
>>>>>>      [phone]
>>>>>
>>>>>
>>>>
>>>>     -- 
>>>>     -----------------------------------------
>>>>     Gunnar Hellström
>>>>     Omnitor
>>>>     [e-mail] <mailto:[e-mail]>
>>>>     [phone]
>>>>
>>>>
>>>>     _______________________________________________
>>>>     SLIM mailing list
>>>>     [e-mail] <mailto:[e-mail]>
>>>>     https://www.ietf.org/mailman/listinfo/slim
>>>>     <https://www.ietf.org/mailman/listinfo/slim>
>>>>
>>>>
>>>
>>> -- 
>>> -----------------------------------------
>>> Gunnar Hellström
>>> Omnitor
>>> [e-mail]
>>> [phone]
>>> _______________________________________________
>>> SLIM mailing list
>>> [e-mail] <mailto:[e-mail]>
>>> https://www.ietf.org/mailman/listinfo/slim
>>
>
> -- 
> -----------------------------------------
> Gunnar Hellström
> Omnitor
> [e-mail]
> [phone]

-- 
-----------------------------------------
Gunnar Hellström
Omnitor
[e-mail]
[phone]