Contributors mailing list archives
Re: Proposal for new workflow, incorporating "Optimistic Merging"by
Willdoo IT, Richard deMeester
Our full support to this proposal. I have personally become reluctant to contribute when the style issues, etc, become onerous.
This is particularly the case when taking an existing module, making it work in version 9, and then having to change and alter all manner of points that were all existing in version 8. I was planning to upgrade another 4 or 5 re partner features, and just backed away.
Let's keep things moving.
On 7 June 2016 at 08:52, Daniel Reis <firstname.lastname@example.org> wrote:
I think we need to step back and look at the bigger picture: The primary goal for the OCA has been quality. That was achieved achieved: we now have a consistent method to publish quality modules. We have done a long way since the Launchpad and pre-OCA days. We won't give up on that. That's not under discussion here. The point is to find ways to make the community grow. Bigger, better, faster, stronger. We may think that allowing for this "bad code" to hang around isa waste of time. But at the same time it provides a stage for learning and experimentation. And motivate new contributors to gradually learn all the the guidelines and best practices. It has been pointed out the lack of reviewers is an issue. I believe that optimistic merging provides answers to that. It changes the criticize-fix cycle, where the burden is mostly on the original contributor, to get everything right by himself, into a fix-fix cycle, where incremental improvement is made. I argue that the question is not *if* we want to do that, but rather *how* it can be done. So let us focus on Solutions and Objections to them, rather than questioning if it can be done or not. And I think that we can experiment. Some people, including me, are willing to drive experiments on a few specific repos. But first we need to agree on what to try. Best, Daniel
ThinkOpen Solutions Portugal, Daniel Reis