On Tuesday, 9 September 2014 21:44:25 CEST, Alexander Neundorf wrote:
Having two different patch review systems for one project... I mean, this is surely not a good idea. Two places to send patches, to places to review patches, two different user interfaces,
This is not a final state. To be honest, I think that having a flag-day migration would be even worse, so running with two competing systems for a while seems like the only plausible way to proceed here -- otherwise one has that annoying chicken-and-egg problem.
It's up to kio's and plasma-framework's maintainers to decide what they want in the long term. I know that it's Gerrit for trojita.git, and I sincerely believe that the alleged UI glitches of Gerrit are far outweighted by the improvements which the whole stack brings, such as having each changeset available as a git branch to pull from. These benefits will only be amplified when it's integrated with a pre-approval CI runs, something which is just impossible to do with RB without a ton of ad-hoc and fragile scripting. I will surely appreciate a system which tells me whether a random patch I'm about to review at least builds, or whether there are any regression in there.
Please note that some of the often-quoted UI and usability glitches have been already fixed in the version of Gerrit we're running, so I would strongly encourage people to actually give Gerrit a try before they dismiss it based on personal experience with another Gerrit instance a couple of years ago. Some quick instructions are at [1].
maybe logins, ...
KDE's Gerrit accepts KDE Identity logins, of course. [1] https://techbase.kde.org/Development/Gerrit With kind regards, Jan -- Trojitá, a fast Qt IMAP e-mail client -- http://trojita.flaska.net/
