I'll drive this. Let me try to wrangle a draft proposal together. Dirkjan replied to me and that feedback is helpful.
On Tue, Jun 4, 2013 at 2:55 PM, Noah Slater <[email protected]> wrote: > Note: I don't mean to put this on Bob. Anyone could drive this. But I do > think it needs a driver. Bob, Randall, and Dirkjan all seem to have the > most detailed thoughts on the subject, so I suggest one you might be in the > best position. > > And to clarify: someone disagreeing with you isn't a blocker. We're aiming > for discussion-lead decision-making. Feel free to supply the > "decision-making" that compliments the "discussion-lead" part. ;) > > > On 4 June 2013 22:36, Noah Slater <[email protected]> wrote: > >> Odd way to phrase it. Alternative proposals should not be >> a destructive part of the process. The goal here is to generate ideas, toss >> out the ones that don't work, pick your favourite, and drive consensus on >> it. >> >> So, there are two possible ways I can see this unfolding: >> >> * Everyone agrees with you that the git-flow stuff is not needed, in >> which case, great. Work everyone's comments in to the original proposal, >> and then move it from DISCUSS to VOTE. >> >> * There is still some disagreement about what we want to do. In this >> case, I agree, we do not have consensus. (I wouldn't describe this as >> a destruction of coherency. Instead: productive discussion!) The next >> step forward in this instances is to drive that discussion, and hopefully >> come out with a proposal that most people like. >> >> I note that Randal posted several mails, and so did Dirkjan. But nobody >> has responded to them. A good way to kick this off again would be to >> respond to those points, I think. >> >> I would love to drive this, but I can't, mostly because I have no idea >> what I'm talking about when it comes to Git. ;) >> >> >> On 4 June 2013 19:03, Robert Newson <[email protected]> wrote: >> >>> Heh, if I felt I could conclude that thread I would have done so >>> already. We had a reasonably well described approach at one point and >>> coherency was destroyed by a late appearance of the git-flow >>> alternative. >>> >>> B. >>> >>> >>> On 4 June 2013 18:43, Noah Slater <[email protected]> wrote: >>> > This thread is concluded. :) I meant the "[DISCUSS] Git workflow" >>> thread. >>> > >>> > >>> > On 4 June 2013 18:41, Robert Newson <[email protected]> wrote: >>> > >>> >> What's not concluded in this thread? >>> >> >>> >> B. >>> >> >>> >> >>> >> On 4 June 2013 18:04, Noah Slater <[email protected]> wrote: >>> >> > Bob, are you able to help drive the Git thread to conclusion? We >>> need to >>> >> > clarify this and document it. Think a lot people are confused right >>> now >>> >> > since it seems everything is in the air. >>> >> > >>> >> > >>> >> > On 31 May 2013 16:51, Robert Newson <[email protected]> wrote: >>> >> > >>> >> >> master, as usual, and the x.y.z branches (for backports). All other >>> >> >> branches should be feature or fix branches we've not deleted. >>> >> >> >>> >> >> B. >>> >> >> >>> >> >> On 31 May 2013 16:47, Wendall Cada <[email protected]> wrote: >>> >> >> > I'm fairly well versed in using git and different workflows, >>> rebase, >>> >> etc. >>> >> >> > However, I'm utterly confused as to how I might contribute >>> changes to >>> >> >> > couchdb, what branches are relevant, etc. Is there documentation >>> for >>> >> >> this, >>> >> >> > or a clear summary of decisions made? >>> >> >> > >>> >> >> > Thanks, >>> >> >> > >>> >> >> > Wendall >>> >> >> >>> >> > >>> >> > >>> >> > >>> >> > -- >>> >> > NS >>> >> >>> > >>> > >>> > >>> > -- >>> > NS >>> >> >> >> >> -- >> NS >> > > > > -- > NS
