Re: [alto] WG Review: ALTO Charter Update(Internet mail)

[Qin Wu]

Hi, Chunshan and All:
With chair hat on, which use cases the WG should pursue is *exactly* the 
intention of the “informational” deliverable to identify any use cases that 
have strong support and a realistic chance of implementation and deployment.

The purpose of that step is to help guide the WG and ADs for a future 
re-chartering activity. It seems we have a choice to do the light-weight 
recharter as proposed or to delay doing any re-charter until
after those use cases have been identified and discussed.

It would be better to be able to move forward with other (small) important ALTO 
work while we work out what we might do next. In the meantime,

it is important that everyone contributes to the implementation and deployment 
survey as well as the detailed discussion of the future implementation and 
deployment of any protocol extensions.

For your comments, please my reply inline (without chair’s hat).

-Qin
发件人: chunshxiong(熊春山) [mailto:chunshxi...@tencent.com]
发送时间: 2021年5月13日 21:58
收件人: Qin Wu ; IETF ALTO 
抄送: alto-cha...@ietf.org; Zaheduzzaman Sarker 
主题: RE: [alto] WG Review: ALTO Charter Update(Internet mail)

Hello Qin,

Thanks for the feedback!

Regarding to the recharter proposal, we attended the discussion quite long time 
(though quite a lot are not the WG level but within design team).  The bullet 
you referenced indeed in our view is just for information and may not end up 
with a WG document.  So this is indeed not our expectation:-)

o Report back to the Area Director to identify any use cases that have strong 
support and a realistic chance of implementation and deployment.

So, regarding to your proposal to spit into different documents, thanks for the 
proposal but we will evaluate and discuss within design team and decide whether 
to continue.
[Qin]: I think you should continue to cook your draft or break down as I 
suggest, a good work starts from use case, requirements, and framework and need 
to undergo a period of development and several iterations of review by the 
Internet community. Sometime take time to get everybody on the same pages.
Many good example use cases and requirements documents you can find in IETF, 
such as RFC6770 (CDNI use case), RFC6708 (ALTO requirements).


Meanwhile, for coordination between 3GPP and IETF, indeed there is official 
LSes always between these two SDOs, but the LS normally is based on the 
progress on standards work related to each other.
The ongoing 3GPP R-18 new study/work item selection sees great interest in ( 
interactive) application network coordination for different kinds of companies. 
If IETF ALTO does not continue to study the new protocol to support the new 
services,
[Qin]: To take on new work, we need to better understand where we are, which 
pieces have been implemented, which pieces are not needed any more, which 
pieces require modification or upgrading.
We need to evaluate whether new requirements can be met by the current 
protocol, what is in the scope, what is not in the scope. To answer these 
questions, we propose two phase work:
phase1 identify what is the solid basis,  phase 2: whether we can use this 
solid basis to support new requirements and new services.
even if 3GPP sends LS to IETF, we may not be able to address it properly. In 
3GPP Rel17, 44 companies from global area support advanced interactive services 
related work ranked as high priority topic during work item prioritization.
 From a service provider perspective, it is also very clear to us how important 
it is to optimize the user experiences for such important and popular scenarios 
like cloud gaming etc.
[Qin]: I think it is your responsibility and every contributor’s responsibility 
to tell IETF and tell ALTO WG What are requirements/Use cases for these 
interactive service? What requirements are specified by 3GPP? what 
requirements/Use cases not and need to be documented in IETF?
How these requirements/use cases are related to ALTO protocol? What are ALTO 
protocol limitation are when you test your use cases? All of these need to be 
built as consensus and follow IETF process and then we can make success story, 
don’t you agree?
 It is really a pity that some of the interested companies
may not come to IETF but it really doesn’t mean such use case are not dominant. 
 In future we can consider to ask more companies to come to IETF or ask their 
3GPP team to align with IETF team:-)  Regarding to new wheels, we think there
is very clear spit/boundary between 3GPP and IETF thus such cases can probably 
be avoided easily.  Anyway, this is something in the future and we can come to 
IETF when proper.

So, again thanks for Qin’s response and we do hope ALTO  is actually addressing 
how state-of -the-art dominating application and network can coordinate to 
improve user experiences.

Thanks a lot!
BRs,

Chunshan Xiong

From: Qin Wu mailto:bill...@huawei.com>>
Sent: Wednesday, May 12, 2021 8:14 PM
To: chunshxiong(熊春山) mailto:chunshxi...@tencent.com>>; 
IET

Re: [alto] WG Review: ALTO Charter Update(Internet mail)

[熊春山]

Hello Qin,

Thanks for the feedback!

Regarding to the recharter proposal, we attended the discussion quite long time 
(though quite a lot are not the WG level but within design team).  The bullet 
you referenced indeed in our view is just for information and may not end up 
with a WG document.  So this is indeed not our expectation:-)

o Report back to the Area Director to identify any use cases that have strong 
support and a realistic chance of implementation and deployment.

So, regarding to your proposal to spit into different documents, thanks for the 
proposal but we will evaluate and discuss within design team and decide whether 
to continue.

Meanwhile, for coordination between 3GPP and IETF, indeed there is official 
LSes always between these two SDOs, but the LS normally is based on the 
progress on standards work related to each other. The ongoing 3GPP R-18 new 
study/work item selection sees great interest in ( interactive) application 
network coordination for different kinds of companies. If IETF ALTO does not 
continue to study the new protocol to support the new services, even if 3GPP 
sends LS to IETF, we may not be able to address it properly.
In 3GPP Rel-17, 44 companies from global area support advanced interactive 
services related work ranked as high priority topic during work item 
prioritization.   From a service provider perspective, it is also very clear to 
us how important it is to optimize the user experiences for such important and 
popular scenarios like cloud gaming etc.  It is really a pity that some of the 
interested companies may not come to IETF but it really doesn’t mean such use 
case are not dominant.  In future we can consider to ask more companies to come 
to IETF or ask their 3GPP team to align with IETF team:-)  Regarding to new 
wheels, we think there is very clear spit/boundary between 3GPP and IETF thus 
such cases can probably be avoided easily.  Anyway, this is something in the 
future and we can come to IETF when proper.

So, again thanks for Qin’s response and we do hope ALTO  is actually addressing 
how state-of -the-art dominating application and network can coordinate to 
improve user experiences.

Thanks a lot!

BRs,

Chunshan Xiong

From: Qin Wu 
Sent: Wednesday, May 12, 2021 8:14 PM
To: chunshxiong(熊春山) ; IETF ALTO 
Cc: alto-cha...@ietf.org; Zaheduzzaman Sarker 
Subject: RE: [alto] WG Review: ALTO Charter Update(Internet mail)

Thanks Chunshan for your input and comments on charter proposal, see reply 
inline.
发件人: chunshxiong(熊春山) [mailto:chunshxi...@tencent.com]
发送时间: 2021年5月11日 16:40
收件人: Qin Wu mailto:bill...@huawei.com>>; IETF ALTO 
mailto:alto@ietf.org>>
抄送: alto-cha...@ietf.org; Zaheduzzaman Sarker 
mailto:zaheduzzaman.sar...@ericsson.com>>
主题: RE: [alto] WG Review: ALTO Charter Update(Internet mail)

Hello WuQin and working group,

I provide our views on this recharter.

Firstly, we think ALTO is an IETF WG to standardize the interaction between 
application and network, initially for P2P application and now it is a good 
chance to continue optimization to support these new interactive services like 
Cloud Gaming, XR/AR, V2X application etc.
[Qin]: I agree to support further evolving of ALTO protocol.
To support new application and introduce further optimization for ALTO 
protocol, we need to get more implementation deployment and experience to help 
us better understand which piece works, which pieces not needed, which pieces 
need to be redesigned. ALTO protocol is initial designed for P2P, later on CDN 
application, it is generic protocol, Do we have P2P specific features that need 
to peel off? To get this question answer, we propose the first work item and 
the second item and will create wiki page to keep track of related 
concern/issues/report

Secondly, we have been working together with colleagues from network operator, 
network vendor and academy et. al. for quite long to perform ALTO-oriented 
research and also real network testing which have already show very clear 
benefits and we contribute MoWIE to this WG 
（https://datatracker.ietf.org/doc/draft-huang-alto-mowie-for-network-aware-app/).
  We think the listed items, i.e. the generic protocol extension for policy 
attributer, proposed as the high priority for recharter proposal in IETF#110 
ALTO shows rough consensus to some extent. If ALTO WG doesn't continue these 
topics explicitly, it is very regretful and we really feel disappointed about 
this.

[Qin]: Please see the latest charter proposal, last work item we proposed based 
on list discussion
https://trac.ietf.org/trac/alto/wiki/v0.5-recharter
o Report back to the Area Director to identify any use cases that have strong 
support and a realistic chance of implementation and deployment.
New use cases documentation is encouraged, especially the use cases for new 
emerging applications as you mentioned, AR, VR, Cloud gaming, which have strong 
support.
For MOWIE, I think the same question applies