RE: AD Time

["Adrian Farrel" <adrian@xxxxxxxxxxxx>]

It should be no surprise that I, too, have seen this debate before.

 

A very large part of the question hinges on what ADs actually spend their time on. How much of an AD-fortnight [1] is spent reviewing document and handling the backwash from those reviews. When on the IESG I found it (surprisingly?) hard to get the IESG to answer that question.

 

But Stephen Farrell raised this in two ways that might be helpful:

 

1. If we only had one AD for each Area, would that AD feel the need to commission additional reviews to make up for the missing second (or third) AD? If so, then let's do that anyway. If not, then why have all ADs reviewing all the documents? In fact, you could argue for releasing at least 5 ADs from all of their reviewing and follow-up tasks.

 

2. Cannot two ADs from the same Area not share the review load with a little discussion before reviewing starts? This might lead to a 65/65 split (rather than 50/50) and some further brief discussion, but surely it could be made to work. [Hint: some ADs have done this in the past.] This argues for releasing each AD from 35% of their reviewing and follow-up tasks.

 

It should be up to the IESG to self-organise! The only limits we should set are on what work we expect the IESG to get done. The constraint we should give NomCom are that we expect the work to get done, and that any restrictions on work-load should not negatively harm the work of an Area or the IESG as a whole. Thus, questions:

- to the candidates about how much work they can achieve and what

   they would 'sacrifice'

- to the continuing ADs about how much work they are achieving and

  what they need from a co-AD

- to the IESG as a whole about where the pinch-points are with current

  and projected work-loads, and so where resources are needed in the

  new appointments

- the community about what work appears to be falling by the way-side

  and what additional tasks need to be picked up.

 

Thanks (and thanking NomCom for the hard job they are embarking on)

 

Adrian

 

[1] This unit is chosen because the document review cycle in the IESG is currently based on a telechat that occurs every two weeks.

 

From: ietf [mailto:ietf-bounces@xxxxxxxx] On Behalf Of Alia Atlas
Sent: 25 July 2018 17:06
To: Ted Lemon
Cc: IETF Discussion Mailing List
Subject: Re: AD Time

 

Arguments about not being able to find enough good AD candidates (are you prepared with your nominations?) who can spend time basically amount to not having a day job convinced of the value.  Explaining the businesses value of the IETF & the role is needed - not assuming that investments aren't possible.

 

Regards,

Alia

 

On Wed, Jul 25, 2018, 11:58 AM Ted Lemon <mellon@xxxxxxxxx> wrote:

No, what I'm getting at is that I think that's the wrong knob to turn.

 

On Wed, Jul 25, 2018 at 12:18 AM, Brian E Carpenter <brian.e.carpenter@xxxxxxxxx> wrote:

On 25/07/2018 15:53, Ted Lemon wrote:
> The problem with review teams is that if you don't read the documents and
> don't know what they are about, you don't have the overview that allows for
> synthesis.   One of the advantages of having people who review "all" the
> documents is that stuff occurs to those people because they see connections
> that people who don't review "all" the documents don't get.   I put "all"
> in quotes because it's never really all, but even so, ADs definitely have a
> bird's eye view that is not shared by anyone else.

It's true. But do you have any other ideas how to *substantially*
reduce the AD workload?

   Brian

>
> On Tue, Jul 24, 2018 at 11:21 PM, Brian E Carpenter <
> brian.e.carpenter@xxxxxxxxxx> wrote:
>
>> On 25/07/2018 01:41, Kathleen Moriarty wrote:
>> .....
>>> I think that as AD my time was consumed because I made a point of
>> reading,
>>> or at least skimming, all drafts prior to publication looking for
>> security
>>> specific issues.
>>
>> So would things be better if we formalized the area review teams so
>> that they perform this function directly and can officially register "No
>> Objection" in the IESG ballot, with the AD only being involved when the
>> suggested ballot is "Yes", "Discuss" or "Abstain"?
>>
>> (We've been talking about AD overload for >10 years, so maybe it's
>> time to actually change something.)
>>
>>     Brian
>>
>>
>