W3C home > Mailing lists > Public > public-w3process@w3.org > July 2018

Re: How we compile agendas for process CG calls

From: David Singer <singer@mac.com>
Date: Fri, 13 Jul 2018 16:13:36 +0200
Cc: W3C Process CG <public-w3process@w3.org>
Message-id: <CA3E9E4F-150F-436F-B56F-C1A1380C7B12@mac.com>
To: Florian Rivoal <florian@rivoal.net>
Hi Florian

perhaps I should do an introduction; what we aim to do is consider, for the most part (a) proposed changes, in the form of Pull Requests (best) or comments in the issues (less good) or (b) significant discussion in the issues. That is, with calls only once per month, we attempt to run IETF-style; discussion outside the meetings, and decisions in the meeting.

For new issues, I aim to try to decide whether we think it’s something we need to take up this year; and to decide if someone is ‘assigned’ to move the conversation ahead.

So, the reason I ran through the whole landscape on this call was two-fold:
* so we all got a survey of it
* so that we all realized that for the most part, this isn’t happening; we’re not doing between-meeting discussion, not getting Pull Requests and (I am as guilty as others) the assigned people are not advancing the state of discussion on their assigned issues.

Indeed, in the past when people had significant reactions to the proposals or discussions that had happened since the previous call, we’ve drilled down into one or two issues. Also, last year we had a couple of things we knew we must finish, and so we spent a lot of time on them.

Maybe this model isn’t working, but I don’t think many of us want to go to a model where we do the work in long weekly calls.

Thoughts?

> On Jul 11, 2018, at 17:26 , Florian Rivoal <florian@rivoal.net> wrote:
> 
> Hi,
> 
> Today was my first time joining the process CG call, and while that makes me a bit of a confused noob, it also gives me the chance to bring some fresh perspective.
> 
> Going through the entire list of open Process2019Candidate issues did not seem to be very efficient to me. For many issues, we ended up saying that no work had happened yet, or that we were waiting for someone's input, or that this should instead be discussed at the AB...
> 
> This takes a lot of time, and leaves relatively little for discussion the issues where we can actually do a deep dive.
> 
> Relatedly, when preparing for the call, the list of issues included in the agenda (via the various links) was very long, which meant that although I did prepare and reviewed the issues, I could only do so superficially for the vast majority of them.
> 
> As an alternative approach, I suggest that we create an "Agenda+" label that participants in the CG can add to any issue they would like to discuss in the next call, and that gets removed once we have actually discussed it (and can be re-added as soon as further time on the call is desired).
> 
> If some of the Agenda+ topics are Process2019Candidate and some are not, we should probably discuss those that are first, and those that are not only if time allows.
> 
> And if we're not Agenda+ing enough Process2019Candidate topics (or not enough topics at all), as the chair, dsinger should absolutely be calling out the cg's members (and particularly AB members) and bugging us to make sure we do work. But mostly this can happen outside the call.
> 
> I think this would allow for more efficient use of meeting time and of prep time, which hopefully translates into faster progress on the process itself.
> 
> —Florian

David Singer

singer@mac.com
Received on Friday, 13 July 2018 14:14:08 UTC

This archive was generated by hypermail 2.3.1 : Friday, 13 July 2018 14:14:09 UTC