Even if it's just a partial move, it'll split review. And I think it's extremely unlikely that we'll manage to very clearly separate what's allowed to happen in github PRs, and what on the list.
-
-
Replying to @AndresFreundTec @fuzzycz and
I have to agree. I think that having an expedited review process for small changes (definition TBD) makes sense, but having multiple streams into the codebase would be a major headache.
1 reply 0 retweets 0 likes -
Replying to @Xof @AndresFreundTec and
Obviously, it's not without risks. I agree changes that are not obviously correct should be funneled to the list, and committers / reviewers should not be expected to watch GH. If there are volunteers willing to curate the GH queue, why not to give it a try?
1 reply 0 retweets 0 likes -
Outside of typo-fixes etc there's just about no obviously correct contribution by first timers. It's possible that those would suddenly appear, but I doubt it. And if there's any discussion, how would we guarantee it's archived somewhere under our control?
1 reply 0 retweets 2 likes -
Replying to @AndresFreundTec @fuzzycz and
And why are we discouraging even the small drive-by patches by making the process unnecessary complicated for people not familiar with the PostgreSQL development process?
2 replies 0 retweets 0 likes -
Replying to @ascherbaum @fuzzycz and
Well, as I previously said. There's very few first comer contributions that actually can just be applied, and the rest requires discussion. And those need to happen somewhere where others have a chance to intervene. Doing that in GH would mandate committers watching it.
3 replies 0 retweets 0 likes -
Replying to @AndresFreundTec @ascherbaum and
Honestly, I think there's somewhat of an infrastructure/process here (i.e. not having http://contributing.md not auto-closing PRs with referral). But the much larger issue is that there's basically no human resources to do that kind of work.
3 replies 0 retweets 0 likes -
Replying to @AndresFreundTec @ascherbaum and
Just adding more committers itself will barely help IMO (if not regress things) - what we need is committers that have this kind of work as a substantial portion of their day job. And then not get harassed / disadvantaged by doing fewer impressive features.
2 replies 0 retweets 2 likes -
Replying to @AndresFreundTec @ascherbaum and
And that requires resources. I think that's a much bigger issue than "allow github". Unless we have those there'll be not resources to merge trivial stuff either. It's not like there's not plenty of that on the lists.
1 reply 0 retweets 1 like -
Replying to @AndresFreundTec @ascherbaum and
And I think it's not just committers where such resources are lacking. There's e.g. very little accurate docs for the development process, and related things. And I think that again is because it's largely unsexy to our employers.
2 replies 0 retweets 2 likes
To be clear: Big ticket features outstripping anything else for companies is one issue. But the generally tiny amount of funding of postgres another. But that's also because we make it just about impossible for companies to do so usefully.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.