Re: Subject: CFV: New JDK 10 Committer: Erik Österlund (original) (raw)

Roman Kennke roman at kennke.org
Thu Jun 22 14:01:30 UTC 2017

Am 22.06.2017 um 15:36 schrieb Claes Redestad:

On 2017-06-22 14:45, jesper.wilhelmsson at oracle.com wrote: No rule change needed, but a social shift is long overdue. I'm not sure what you mean by social shift. Do you mean that we should lower the bar for what a significant change is? Or do you mean that we should let our friends slip by because we know that they are good developers? I think your bar for what's considered a significant change is generally way too high, yes, especially for a Committer rule.

I agree with that. Another way to look at it is to ask 'have we ever run into problems because the barrier of entry is too low?' I'm thinking the sort of problem where some random guy commits a fix or two, then disappears and leaves others with a trail of bugs. I am not aware that something like this ever happened to OpenJDK. On the other hand: 'did we ever get problems because the barrier to entry is too high?' Probably. We it not for the army of paid developers, OpenJDK would most likely look pretty poor as an open source project. And I can remember a bunch of occasions were competent volunteers with great patches almost fell to the wayside because of all the bureaucrazy involved.

I fully agree that the rules should be changed, mostly to take out the wording of "significant" from the rules (instead explicitly enumerate what types of changes are not to be counted), but merging the Author and Committer roles would also be welcome. It would be very useful to spell out the purpose of the rules. What do we want to achieve by having those rules? This would probably help to not get caught up and waste time in arguments about strictly following the rules, but do something useful for OpenJDK instead.

Roman

More information about the jdk10-dev mailing list