Dear Martin,
-----Original Message-----
From: Martin Stiemerling
[mailto:martin(_dot_)stiemerling(_at_)neclab(_dot_)eu]
Sent: Tuesday, September 18, 2012 4:36 PM
To: Songhaibin
Cc: Richard Woundy
Subject: Re: DECADE WG to be closed
Dear Haibin,
On 09/17/2012 11:39 AM, Songhaibin wrote:
Dear Martin,
Hope everything goes well with you and thank you very much for your efforts
to reviewing the drafts in detail and giving guidance.
As I agree with most of your comments to the DECADE requirements draft, but
I have to say IMO the architecture document is not that bad. This document
gives
a clear description of the DECADE server/client components and
implementation/design principals which will be reflected in the protocols, IMO
this is what an architecture document should do.
I do not agree there is lack of technical substances to design a base
protocol
which can satisfy the transport and resource control requirements for content
distribution applications. Some detailed design choices are still not very
clear, and
need efforts for them.
And recently, the energy is growing, we recently received a lot of list
discussion
including comments from Kostas about the requirements and architecture and
also a new individual draft for the service discovery was submitted.
The energy has indeed grown in the WG since before the summer. But, I
indicated in my email from mid of June that I have doubts on the
technical quality of the DECADE drafts. These doubts have turned into
certainty, i.e., see the my AD reviews of the requirements and the
architecture.
The technical quality of the drafts would be ok, if the WG would be at
the beginning of the process of discussing and writing those drafts, but
it is not acceptable at the end when the drafts are intended to become RFCs:
The technical base is just to weak to continue from, even after spending
time and effort of the WG participants for more than 2 years.
The requirements document was accepted on Oct. 18, 2010, and the architecture
draft was accepted on March 7, 2011.
Another important data point, as mentioned earlier:
There has been public feedback from IETF community members, such as Dave
Crocker and Carsten Bormann, which questioned the technical base of
DECADE as a whole. This happened at the end of the 2011 and in the first
quarter of 2012.
The was no and still has not been an adequate response from the DECADE
WG to these reviews. For instance, the requirements did get a lot of
feedback from Dave Crocker, but this feedback was never addressed in an
email. I also have been unable to sort out what parts of the feedback
has been addressed in the updated draft and how, and what parts have not
been addressed.
I believe all those comments were addressed in the current draft, as I joined
the discussion with the authors to address the comments. Their efforts should
be respected. The authors and I would like Dave and Carsten to check the draft
with their comments, if they are interested. While I admit answering in the
mailing list is a main method to resolve comments, but it is not the only
method.
I have also received much stronger feedback about the DECADE WG in
private emails to me. Again from long standing IETF community members
that send me feedback arguing that DECADE is not having a technical base
to build on top of.
OK. But general rule for IETF is rough consensus, not private emails. Why not
discuss their questions in the list?
You have asked in your other email to give more time to the WG until the
next IETF meeting in November. This would be one possible way forward,
but I do know about the past 6 months after the IETF meeting in Paris.
Not a lot has happened during this period in order to improve the WG
drafts, in the sense that there is a solid technical base where DECADE
could continue to work from.
I can answer If your question about the technical base can be more specific.
Even if you and the whole WG would start to work full-time on the
drafts, it still would take longer than to the next IETF meeting to move
the requirements and architecture forward. My gut guessing is that it
will take at least until March 2013.
To give an example:
It is completely unclear how the resources on a DECADE server are
supposed to be managed and how this management is mapped to the protocol
split of SDT, DRP, and other management protocols.
Parts of it, such as setting the permissions of data objects clearly
belongs to the DRP, and it is sort of stated in a vague way in the
architecture, but it is not documented in a comprehensive way. Other
parts, such as the accounting is probably not part of the the DRP nor
SDT, but there is supposedly another interface that is needed for this.
Has this been discussed at any point in the WG?
I just read the email that Richard answered these questions with text from the
current drafts. And I agree with his answers.
While I respect that AD can make the decision of closing a WG, but I see a
dozen of emails expressed their disappointment.
BR,
-Haibin
Given the above points and my summaries out of the last email and the
one of 6/12, the DECADE WG is going to be closed by today.
The DECADE WG mailing list will remain open until the end of the year,
to let the people a chance to discuss how to go forward with the drafts.
As suggest in my earlier email:
The participants are free to overhaul the drafts and to submit them as
individual submissions to the RFC Editor's Independent Stream.
The decisions to close the WG can be of course appealed via the IETF
appeal process:
See 'Appeals and PR-Actions' under http://www.ietf.org/iesg/ and RFC 2026.
Martin
BR,
-Haibin
-----Original Message-----
From: Martin Stiemerling
[mailto:martin(_dot_)stiemerling(_at_)neclab(_dot_)eu]
Sent: Friday, September 14, 2012 7:53 PM
To: Songhaibin; Richard Woundy
Subject: DECADE WG to be closed
Dear Rich and Haibin,
I have finally done my AD review for the DECADE architecture draft after
finishing the DECADE requirements draft.
The first feedback for the DECADE architecture draft has been provided
in the datatracker and sent to the authors and you by email.
Both drafts are in an extremely bad shape, i.e., they would require a
major overhaul and have been sent back to the working group due to lack
of technical quality.
I have already expressed my concerns about the energy and the lack of
technical confidence in the group in my summary email of 6/12. The
requirements and architecture drafts got advanced towards the IESG
afterwards. The push for energy was good.
However, after reviewing the two key drafts, requirements and
architecture, and receiving feedback from IETF community members, I have
come to the conclusion that the DECADE working group lacks a sound
technical ground.
The DECADE group started its work in end of April 2010 and is now
working for more than 2 years on the milestones/drafts. The time isn't a
big deal, but after 2 years I would have expected that the documents are
on a good technical level where the WG can build on top of.
The issues for the potential future protocol works is that if the basics
are not well understood and documented, how can the protocols be
designed in a comprehensive and technical sound way?
I cannot see this anymore.
This was also documented in my email on 6/12:
"
I have seen reviews for the ps, the reqs, and the architecture drafts
which go all in the same direction: where is the technical substance,
DECADE will built on?
The last meeting in Paris was really discouraging with respect to the
technical substance...
Yet another sign of lack of energy in the WG...
"
The WG did get a grace period starting after the IETF meeting in Paris
and had the chance to really show that it is moving in the right
direction. However, the current state does still not document this and
therefore the DECADE WG will be closed in the next week. I will inform
the WG on Tuesday afternoon CEST.
The draft authors of the requirements, architecture, and also the
Integration Examples of DECADE System can submit the respective drafts
via the Independent Stream of the RFC editor (see
http://tools.ietf.org/html/rfc6548 for further information), if they
wish to.
Regards,
Martin
--
IETF Transport Area Director
martin(_dot_)stiemerling(_at_)neclab(_dot_)eu
NEC Laboratories Europe - Network Research Division NEC Europe Limited
Registered Office: NEC House, 1 Victoria Road, London W3 6BL
Registered in England 283
--
martin(_dot_)stiemerling(_at_)neclab(_dot_)eu
NEC Laboratories Europe - Network Research Division NEC Europe Limited
Registered Office: NEC House, 1 Victoria Road, London W3 6BL
Registered in England 283