I echo Dave's recommendation to be inclusive. The main reason not to be
exclusive AIUI is to prevent gaming of the voting process. There are two ways
to address that, while remaining inclusive:
1) set a reasonably low bar for participation
IMO anyone that is a regular attendee in OPNFV TSC or technical community
calls, has made any level of peer-reviewed commit (outside their own company)
over the last year, etc should be included. There are a core of people that
should be obvious to us all, as involved, and they should certainly not be
excluded. These include all current TSC members, PTLs, active committers,
anyone active in any form of discussion, anyone contributing content in any
form (code, tests, docs, wiki, ...), ...
2) provide a means for process concerns to be raised and addressed by the TSC
Dave's concern is an example. We need to be open to any such concern as a sign
the process needs to be improved or an exception needs to be made (which really
means the process needs to be improved, but for some reason we are not able to
at this time...). Other types of concern may be raised by analytics on the
voting process, which should be shared with the community (every voter should
be associated with a member company, or identified as independent, so we can
ensure reasonable equity in the voting process). There should be a place on the
wiki etc for raising these concerns so they can be tracked and the thread of
addressing them is recorded.
Bryan Sullivan | AT&T
[mailto:opnfv-tsc-boun...@lists.opnfv.org] On Behalf Of Dave Neary
Sent: Wednesday, February 14, 2018 5:49 AM
To: Raymond Paik <rp...@linuxfoundation.org>; opnfv-tech-discuss
Subject: Re: [opnfv-tsc] [opnfv-tech-discuss] For the TSC composition
discussion on Tuesday
[Note post TSC call: this is the email I had written and found in my
Drafts folder after the TSC call - so I had not sent it - DN]
Thank you for putting this together, Ray!
A few comments: Your wiki stats look off - I would expect to see many
more people in the list (of course the first thing I did out of vanity
was look for myself, and I have definitely made a number of wiki edits
and comments, but I am not in the list).
We have so far discussed erring on the side of inclusion, so I am
curious about your setting a bar at 50 or 100 contributions. It might
make sense to have a minimum number for some of the lower impact
activities like Gerrit reviews, but for others like patch submission, a
lower bound of 1 might make more sense. For wiki edits 5 or 10 seems
reasonable. If using a composite metric, I would lean towards a lower
number (say, 20) rather than higher, to be more inclusive.
Have you considered being active on the mailing list as a potential
market of activity? Again the question of whether people who are active
on the list, but inactive elsewhere, can be considered active
contributors (I think they could) - there, perhaps 30 emails during the
year is a good level.
I would also be interested to hear if there are people who previously
had a vote as committers, who would not have a vote under this scheme,
or whether there is a big difference in the size of the
community/electorate with your proposed levels.
What do you think?
On 02/12/2018 01:12 AM, Raymond Paik wrote:
> This is for the TSC composition discussion on Tuesday.
> As was discussed previously
> there was a consensus to look at a "union of contributions" across
> various tools in OPNFV including Git, Gerrit, JIRA, and Confluence. For
> example, we talked about people making a total of 50 or 100
> contributions across all tools over a 12 month period as the constituent
> for the TSC election.
> In the attached, you'll see the data point across the 4 tools in 2017.
> In the last tab, you'll also find a comparison of "top 50 contributors"
> across the tools. Although there are some exceptions, you'll see that
> active contributors are active across all 4 tools. One of the concerns
> was that we want to be inclusive to recognize non-code contributions and
> you'll see a high number of non-code contributors in both Gerrit and Jira.
> In terms of a threshold, 100 annual contributions seems like a good
> starting point. As a point of reference, the following shows the number
> of people that made 100 or more contributions in each tool. (Based on
> this, we'll have a minimum of 112 people eligible for the TSC election
> as we have 112 people that made 100 or more contributions to Gerrit alone)
> * Gerrit: 112
> * Git: 30
> * JIRA: 36
> * Wiki: 4
> If we go to 50 annual contributions, I don't necessarily think there'll
> be a significant increase in the pool and following is the breakdown.
> * Gerrit: 137
> * Git: 51
> * JIRA: 62
> * Wiki: 8
> Please feel free to reply with any thoughts or feedback. This will be
> discussed further during the TSC call.
> opnfv-tech-discuss mailing list
Dave Neary - NFV/SDN Community Strategy
Open Source and Standards, Red Hat -
Ph: +1-978-399-2182 / Cell: +1-978-799-3338
opnfv-tsc mailing list
opnfv-tech-discuss mailing list