Re: [ieee-ietf-coord] requesting an EtherType

[ROBERT GROW <[e-mail]>]

Dan:

Your suggestion is consistent with what the RAC has encouraged both IEEE-SA projects and the RA to do on Ethertype, Group Address and similar assignments for IEEE standards.  We resist providing the assignment early in the process, but are happy to make the assignment for inclusion in the draft for initial IEEE-SA Sponsor ballot.  Up to that point the RAC prefers the value to be listed in the draft in a way indicating the value will be assigned for Sponsor ballot.  Perhaps we have done a poor job in communicating that concept to the IETF for inclusion in the draft at a similar point of stability.  Perhaps a little more formal process description will help to spread the knowledge a bit broader. 

—Bob



On Sep 25, 2016, at 8:21 AM, Dan Romascanu <[e-mail]> wrote:

> Bob, 
> 
> Thank you very much for your detailed answer, which helps to clarify many issues. 
> 
> I have only two comments. 
> 
> 1. The IETF definitely respects and appreciates the challenges in managing number spaces, and the process put in place by the RAC to best administrate the spaces under its responsibility. We face similar issues with numbers under our responsibilities, and you may be sure that the IETF WGs and the IESG make all that is possible and we know to do in order to avoid submitting frivolous applications. 
> 
> 2. This being said, mistakes can happen, and applications may not include from the beginning all required information. What I would suggest that we do in the spirit of cooperation between the IETF and the IEEE 802 is to avoid as possible the errors and at the same time do best effort to shorten processing time. That is why when a document is sent to the ieee-ietf coordination list with a clear indication that it includes some kind of request for assignment from the RAC, it would be useful to be reviewed as early as possible - desirably at WGLC or IETF LC, and not after the document approval by the IESG (which is roughly equivalent to passing an IEEE Sponsor Ballot). Also, after the application is submitted, it would be good to have some confirmation that it was received and is under processing, and when possible an answer earlier than three months. Note that this time is from the IETF publication process a complete freeze in waiting for the answer, which impacts the 'time-to-market'. 
> 
> Regards,
> 
> Dan
> 
> 
> On Sat, Sep 24, 2016 at 11:18 AM, ROBERT GROW <[e-mail]> wrote:
> IETF folk will appreciate the challenges that occur as a result of exhausting a number space.  The Ethertype field does not provide a huge number space and consequently, conservative assignments for new applications is considered important.  Avoiding the need to come back for additional assignments is the major focus of what the review process looks for.  These design items could certainly be examined early in the development of a draft.  Highlighting that the protocol was designed with these considerations will accelerate the review of an application.
> 
> The Ethertype tutorial (http://standards.ieee.org/develop/regauth/tut/ethertype.pdf) asked some of the questions that will be asked if not answered on the application.  This includes use of a previous assignment, with its sub typing capability, and certainly designing the new protocol to include sub typing.  Simply pointing the Registration Authority (RA) to the draft isn’t going to be as easy to review as an application that either explains the important characteristics of the protocol, or points to where in the draft the information can be found.  
> 
> 1.  For example, explaining that the protocol was prototyped using IEEE Std 802-2014, sub clause 9.2 answers most of the questions that would be asked about the protocol (with the assurance the the sub typing illustrated in that standard’s Figure 12 has been preserved).
> 
> 2.  Absent that using that familiar format for subtype information, help the reviewer understand how similar capabilities are provided.  (E.g., how versions of the protocol will be identified, etc.)
> 
> 3.  Explaining that the protocol is unrelated to other IETF protocols previously having received an assignment is helpful.
> 
> 4.  Knowing the status of the protocol development/standardization is helpful, and can help with timely application review.  There will be reluctance to make an early assignment (e.g., we want to design a protocol to …); because we want a high probability of the protocol being “real” rather than burning an Ethertype assignment for something that may not go anywhere.
> 
> The application review isn’t supposed to be a protocol review (though if a reviewer see a possible fault, they may point that out).  The IEEE RA and the RAC certainly respect the consensus process of the IETF.  Just convince the RA that the application isn’t frivolous.
> 
> —Bob
> 
> 
> On Sep 24, 2016, at 7:03 AM, Dan Romascanu <[e-mail]> wrote:
> 
>> Can we maybe try to cut a little bit from the clerical process and make the earlier review of such requests part of our coordination, at least in what concerns the technical aspects? The Internet-Drafts were written with the purpose of providing the required technical documentation. They went through the IETF consensus process, with its principal milestones (LCs) also announced on this (ieee-ietf) list. I would suspect that if any important information was missing or the request was not targeting the right level, this would have been already flagged up. If something was somehow missed, we  should know earlier than the three month typical response time.
>> 
>> Thanks and Regards,
>> 
>> Dan
>> 
>> 
>> On Sat, Sep 24, 2016 at 6:07 AM, Donald Eastlake <[e-mail]> wrote:
>> I have limited network access right now.
>> 
>> As I recall, you have to provide technical documentation (can reference an Internet draft) which should mention sub typing, etc. (There have been cases of tags where subtyping is not required but it's more difficult to get through the system.) Generally there has been a contractor that examines the application.
>> 
>> If you want to know if your application is in process, I recommend picking up the phone during business hours in the eastern US time zone and calling the office that handles this. You will get a clerical person but at least they can confirm that the application was received.
>> 
>> I believe Pat Thaler has volunteered to informally assist but she may be very busy today and might not get to email for a day or two.
>> 
>> Donald 
>> from iPhone 
>> 
>> 
>> On Saturday, September 24, 2016, Dan Romascanu <[e-mail]> wrote:
>> Hi Bob,
>> 
>> Is there any way to accelerate this process and have responses or possible further clarification questions in a shorter time than the maximal 90 days?
>> 
>> Thanks and Regards,
>> 
>> Dan
>>  
>> 
>> On Fri, Sep 23, 2016 at 3:05 PM, ROBERT GROW <[e-mail]> wrote:
>> It is important to answer the questions on the form clearly, especially about sub typing, including indication of why a new Ethertype is needed versus using an existing IETF assignment.  If the application has already been submitted, and not sufficiently answered, you might get a request for more information.  The application is processed within 90 days.
>> 
>> Bob Grow
>> Chair, IEEE-SA Registration Authority Committee.
>> 
>> 
>> On Sep 23, 2016, at 2:29 PM, Ralph Droms <[e-mail]> wrote:
>> 
>>> My recollection is that suresh has taken responsibility here and submitted the request.
>>> 
>>> - Ralph
>>> 
>>> On Sep 23, 2016, at 3:24 PM, Donald Eastlake <[e-mail]> wrote:
>>> 
>>>> Last time I did this, you just go to the public Ethertype request site, fill out the form saying it is for standards use, indicate you are going to pay by wire transfer, and interact with the IEEE office that handles this and have the invoice amount set to zero.
>>>> 
>>>> Donald
>>>> from iPhone 
>>>> 
>>>> On Friday, September 23, 2016, Alia Atlas <[e-mail]> wrote:
>>>> Hi,
>>>> 
>>>> A document, draft-ietf-forces-interfelfb-06, was approved by the IESG for publication and is in the RFC Editor queue.
>>>> 
>>>> I had assumed that the EtherType request would be made as part of the processing, as per https://www.ietf.org/iesg/statement/ethertypes.html.  That appears to not be the case.
>>>> 
>>>> What is the correct procedure for officially requesting the EtherType so that this document can finally be published?
>>>> 
>>>> Thanks,
>>>> Alia
>>>> 
>>>> 
>>>> -- 
>>>> Sent from Gmail Mobile
>>>> _______________________________________________
>>>> ieee-ietf-coord mailing list
>>>> [e-mail]
>>>> https://www.ietf.org/mailman/listinfo/ieee-ietf-coord
>>> _______________________________________________
>>> ieee-ietf-coord mailing list
>>> [e-mail]
>>> https://www.ietf.org/mailman/listinfo/ieee-ietf-coord
>> 
>> 
>> _______________________________________________
>> ieee-ietf-coord mailing list
>> [e-mail]
>> https://www.ietf.org/mailman/listinfo/ieee-ietf-coord
>> 
>> 
>> 
>> 
>> -- 
>> Sent from Gmail Mobile
>> 
>> _______________________________________________
>> ieee-ietf-coord mailing list
>> [e-mail]
>> https://www.ietf.org/mailman/listinfo/ieee-ietf-coord
> 
> 
> _______________________________________________
> ieee-ietf-coord mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/ieee-ietf-coord