On Tuesday, 9 September 2014 17:39:54 CEST, Aaron J. Seigo wrote:
Would it not make more sense to trial it using newer / smaller / unstable projects, as it is an experiment?
Yes, which is why trojita.git was dogfooding Gerrit before I announced this.
As it stands with plasma-framework in particular, there is now a difference in workflow depending on what *part* of plasma one is working on (framework or workspace). So not only is it now different from the majority of frameworks, it is also "different from itself".
Yes, I understand that this is suboptimal from the consistency point of view. But at the same time, the impression I got at the BoF was that people in general like Gerrit and would like to at least consider adapting it. In order to reach that goal, using it for at least a couple of repositories which are also used by these people who are interested in its ultimate adoption makes a lot of sense, IMHO, which is why we decided to give it a try with kio.git and plasma-framework.git.
Please note that this is "testing" as in "we'll see if people like the setup". The tool itself and its integration with the rest of the KDE infrastructure, the git hooks and what not have all been tested already. I would not dare to propose testing of *that* on somebody else's repo, of course.
That this doesn't follow current documentation (such as it is) for new developers certainly can't help any.
If the test turns out good, the docs will be updated. I do not think that it would be appropriate to update the documentation at this point, though, because Gerrit is still an alternative and not *the* way of submitting patches for these two repos. What I'm looking for now is a certain level of buy-in from the project maintainers. Once these guys are confident that our particular configuration of Gerrit is good, then we can go ahead and it a default for "non-core developers", and that would be the time to update the docs.
+1 for striving to improve KDE's infrastructure[1], -1 for experimenting with significant core repositories that are part of a larger stable release.
I feel that this might be the core of your argument, actually. Do you think that there are any risks with having Gerrit enabled? If so, what are these risks that you see?
[1] even if I have my personal doubts w/regards to gerrit's appropriateness for KDE
Could you please elaborate on this one? With kind regards, Jan -- Trojitá, a fast Qt IMAP e-mail client -- http://trojita.flaska.net/
