[rfc-i] Developing consensus, episode 2: Expected hours of work for the RSE
Glenn at riveronce.com
Tue Feb 8 07:54:32 PST 2011
On Feb 8, 2011, at 8:18 AM, Olaf Kolkman wrote:
> On Feb 8, 2011, at 12:16 AM, Paul Hoffman wrote:
>> On 2/7/11 1:56 PM, Craig Partridge wrote:
>>>>> If you can get the people who made your impression to come forward with
>>>>> their estimates, that would help this discussion a lot.
>>>> Maybe, but presumably if they are still interested in the topic
>>>> they are here and can speak up.
>>> I, like Brian, talked with various folks. In my case, I talked them as
>>> part of developing the original RSE job description and as part of the
>>> interviewing process. I can't name them but I can tell you I communicated
>>> in some form with five folks, whose estimates roughly were:
>>> A: half time
>>> B: half to 3/4 time
>>> C: a bit over half time
>>> D: #'s close to Paul's
>>> E: half time
>>> You can believe me or not. I realize this is not super satisfactory but
>>> like Brian, I can't reveal details unless folks step forward -- note that
>>> some are not on this list.
>>> That said, I think a more compelling answer would be to take the estimates
>>> of folks who have done something close to this job, namely Bob Braden and
>>> Glenn, rather than guesses from those (including me) who have not done it
>>> and are translating from editorial jobs they've done elsewhere.
>> Craig, it is not a question of believing you and/or Brian: I don't think anyone is suggesting bias or dishonesty. Instead, as you say above, your estimates were gotten for a job description that is very different than what we have in front of us now. The numbers I estimated were for the current job description (from Olaf's draft), not the one from 18 months ago, and not from the one that Bob worked under, nor even the one that Glenn worked under.
>> If people here are OK with estimating the hours needed for the RSE based on different job descriptions than what we have now, that might work. However, it also might lead to the same bad situation we had before where the people doing the selecting and the interviewees are working from very different assumptions.
> Your estimates where 15 hrs per week, correct? In the spirit of using the best bits for estimating 0.5 FTE sounds about right, doesn't it?
> What I think is more important is that that the hours cannot be evenly spread but there are 80-100% peaks (e.g. during IETF weeks, or when the person starts the job).
> Would the following description work:
> "The job is expected to take on average half of an FTE (approx 20 hrs per week) whereby the workload per week is expected to be near full time during IETF weeks, be over 20 hours per week in the first few months of the engagement, and higher during special projects."
Several people have recently asked me to comment on this thread.
Olaf's characterization is consistent with my experience of the job as practiced and my analysis of the role going forwards. The minimum required "just to show up", remains roughly 1/3 (37%) time. Even so, the work load depends on the ebb and flow of issues arising from Editor staff (RPC), inter-stream conflicts, and general community debate. The basic work load could tend toward exceeding this level. Beyond "showing up", development projects and related discussion will take additional work, which could be considerable, and which might be time-critical in some cases.
A more detailed analysis can't reliably project the job as needing fewer hours per week on the average. This sort of 'accuracy' just isn't available in this sort of work situation. On the odd chance that during some period "showing up" won't require so much work, there will be more time for development activities.
> Olaf M. Kolkman NLnet Labs
> Science Park 140,
> http://www.nlnetlabs.nl/ 1098 XG Amsterdam
> rfc-interest mailing list
> rfc-interest at rfc-editor.org
More information about the rfc-interest