[rfc-i] Developing consensus, episode 2: Expected hours of work for the RSE

[Paul Hoffman]

On 2/4/11 12:43 PM, Bob Braden wrote:
> Sorry I have been too busy to pay careful attention to this discussion,
> but you are evidently talking about a lighter-weight RSE than I had
> envisioned from earlier discussions and RFCs. Your RSE is essentially a
> bureaucrat, attending meetings, reading (but not responding to mailing
> lists, and writing reports. I thought there was more to the job than that.

I based my estimates on Olaf's proposal in draft-kolkman-rse-2011-00, 
which is not "lighter-weight" or "heavier-weight" but "more 
responsible". He/she would manage, not be a bureaucrat.

> Some detailed comments below.
>
> Bob Braden
>
>
> On 2/3/2011 10:20 AM, Paul Hoffman wrote:
>>
>> This topic is still pretty vital to being part of the description, so
>> I'll toss in my own proposal, and am interested to hear different
>> ones. I structured these around the same organization that Glenn did
>> earlier.
>>
>> Meetings:
>> RFC Editor team weekly meetings, plus prep: 8 hr/month
>> IAB bi-monthly meetings, plus prep: 5 hr/month
>> IETF meetings: 24 day/year, or 16 hr/month
>> Community:
>> Expected major projects, each taking 60 hours of dedicated time,
>> two per year: 120 hr/year, or 10 hr/month
>> Passive following of rfc-interest, ietf-general: 5 hr/month
>
> You're kidding, right?

No.

> How about an hour a day, or 20 days per month,
> for the last item?

I have reviewed the past few years of rfc-interest and ietf-general for 
RFC-related topics.

For ietf-general, when an RFC-related topic comes up, it is almost 
always one of the topics that will be addressed in a "major project": 
format of RFCs, editing guidelines, searchability of the repository, 
better tracking tools. A response from the RSE of "we { are | will be } 
discussing that on rfc-interest" should not take an hour a day.

For rfc-interest, the topics that are not in the "major projects" 
category are almost exclusively for the Production Center. I see nothing 
in section 2 of draft-kolkman-rse-2011-00 that makes me think "the RSE 
is supposed to speak for the Production Center" or "the RSE is supposed 
to make all the small changes that are acted on by the Production 
Center". Sandy/Alice/{SomeAMSite} can speak to the non-policy questions 
that come up about the Production Center.

I recognize that this is quite different than when ISI had the RFC 
Editor contract, or even when you were the post-ISI RSE; however, that's 
the way it looks from the draft. Olaf: if Bob is right and the RSE is 
supposed to be doing much more commenting on Production Center 
discussion, by all means say so and reflect that in the -01 of your 
draft. This would definitely increase my estimate of how much time it 
takes to track rfc-interest because it changes it from passive to active.

>
>> Lightweight following of interactions between authors
>> and the Production Center: 8 hr/month
>
> Probably low, in my experience.
>
> I suspect this is low.

Sure, but that was under the model of more direct management of the 
Production Center. Again, I was basing this on the discussion and Olaf's 
draft and the discussion in Glenn's "observations" draft. To me, this 
goes along with trusting the contractor more and not hovering over all 
their interactions, but watching for hot spots or bad trends. If folks 
want the RSE watching more, this number is easy to increase.

>> Reporting:
>> Monthly summary of Production Center numbers: 4 hr/month
>> Annual summary review of Production Center and Publisher:
>> 12 hr/year, or 1 hr/month
>> Contract reviews:
>> Prepare SOW, help prepare RFP, help interview candidates:
>> 48 hours every 2 years, or 2 hr/month
>>
> Again, you are way off. SOW, RFP, interviewing sounds like 2 weeks (80
> hours)
> every two years.

Could be, depending on how much change is desired at each renewal. Your 
higher estimate adds 32 hr/year or .6 hr/week to my estimate.