[postgis-devel] Committer Guidelines review

[Sandro Santilli]

On Tue, Oct 29, 2019 at 11:16:28AM +0100, rmrodriguez at carto.com wrote:
> Do we still want to keep the authors.git file?

Most probably not. I dropped it with commit 188c3573e

> > Never commit new features to a stable branch without permission of
> > the PSC or release manager. Normally only fixes should go into stable
> > branches.
> 
> I now this isn't new at all, but I've been breaking this point a lot
> since I've never asked for permission. Do we really need it?

Nah, let's keep some cowboy style please.

> > spike branch (those in the spike/username area are to be used for experimentation or for major code refactoring that will destabilize the master After such experimentation is deemed stable, this can then be merged into the master after approval from PSC members
> 
> Do we still have those?

I'd drop the whole spike thing, finding a place to publish
a git branch is easy these days.

> > A list of all project committers will be kept in the main postgis directory in a file called (CREDITS) listing for each committer. This will be the responsibility of the PSC to ensure:
> 
> Not true any more, they are in
> http://postgis.net/docs/manual-dev/postgis_introduction.html#credits

Confirmed.

> > It is the responsibility of the document lead to ensure when new enhancements are added or breaking changes are made, these are noted in the trunk/NEWS or relevant branch/NEWS as soon as conveniently possible. The note should include the trac # (unless a major feature with many tickets) and contributors to the feature/change.
> 
> I think this should be the responsibility of the commiter and done at
> the same time the changes are pushed.

Agreed. Eventually we'll need a policy to make NEWS editing
less conflicting (there are a number of projects out there
to help with that, like adding NEWS items as files in a directory
and then merging them all (and removing) at release time.

--strk;