Re: [manet] DLEP-14: Credit Windowing (Section 9)

["Wiggins, David - 0665 - MITLL" <[e-mail]>]

I think the credit windowing extension needs significant work.  I have
looked briefly at implementing it and I felt like I had to do a lot of
guesswork/assumptions to make sense of it.  I'm concerned that fixing this
will delay the main DLEP draft.  So, I'm in favor of splitting them.

David

From:  Justin Dean <[e-mail]>
Date:  Thursday, June 25, 2015 at 10:42 AM
To:  John Dowdell <[e-mail]>
Cc:  "[e-mail]" <[e-mail]>
Subject:  Re: [manet] DLEP-14: Credit Windowing (Section 9)

I'm going to +1 for keeping it as one draft.  I see merit in breaking it up
into two (and that may still happen) but I also see the value in having it
within a single draft.  It allows reading DLEP for the first time to see how
to go about writing an extension without having to know about another draft
RFC.  We did something similar for SMF and it worked well there too.  Unless
there is further comments on list supporting breaking the document up I
don't think rough WG consensus has changed on this issue just become more
rough.  

Others please comment one way or another as the editor team is working hard
to rev the document and assuming consensus isn't nearly as good as knowing.

Justin Dean

On Thu, Jun 25, 2015 at 9:57 AM, John Dowdell <[e-mail]>
wrote:
> 
> 
> On 25/06/15 14:22, RJ Atkinson wrote:
>>> On 25  Jun 2015, at 09:10 , Ratliff, Stanley <[e-mail]> wrote:
>>> 
>>> The option of splitting the Credit Extensions out into a separate document
>>> was thoroughly discussed, both on-list, and in the DLEP design team. The DT
>>> closed work at IETF 89 (London), reaching consensus on this and other
>>> issues.
>> Stan,
>> 
>> - Design Teams do not get to make decisions for WGs.
>> - Design Teams propose items to a WG for WG review.
>> - So the Design Team discussion above is a bit of a Red Herring,
>>    from an IETF process perspective.
> Ran,
> 
> I've only been at IETF meetings for a few years, and was a member of the
> design team for DLEP. Your comments are noted, but I've never actually seen a
> working group take a decision in the way that a business project team would.
> In practice, working group members either seem to care or not, and if they
> care they usually seem to post regularly on the topic, or are invited to
> become part of a design or author team.
> 
> The design team in this case was a group of individuals who had all regularly
> posted to the list on this topic for a good two years prior to the formation
> of the DT. Despite asking frequently, both on-list and off-list, including
> radio manufacturers and system integrators that I would occasionally meet, few
> other opinions were ever expressed on-list. Many members of the design team
> have implementations, which yielded valuable experience in what worked and
> what didn't, and this went back into the draft. We're now four and a half
> years since draft-00. I know there are major organisations waiting for DLEP to
> get finished so they can inform their own strategies and development cycles,
> and for that they need a stable and implementable specification that those of
> us who have a stake in this protocol can agree on.
> 
> I don't for a second dispute your right to make suggestions or propose text or
> provide review based on implementation experience with this or a similar
> protocol, but those who do care about this draft have lived and breathed it
> for at least the four and a half years it has already existed. I would just
> ask you to consider your comments in the light of this situation.
> 
> John
> 
> 
> _______________________________________________
> manet mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/manet