On Wed, Feb 5, 2025 at 7:14 PM Peter Geoghegan <pg@bowt.ie> wrote:
> Evidently this new policy is why my skip scan patch series wasn't
> being tested by CI.
It happened again, this time with the index prefetching patch that
Tomas and I are working on.
I checked the CF app entry, but didn't notice that the entry wasn't in
the current CF. It was in 2025-11-01 – 2025-11-30, but technically
when I posted the patch that was already over (not in my time zone, in
UTC). Technically I hadn't submitted the patch to the next open commit
fest yet, and yet there is no real indication that that's a problem on
the CF page.
I thought it was a bug in the CF app, and then complained about it on
Discord. Then Jelte wasted a couple of hours of his own time on this,
before he finally noticed that the patch just wasn't in a CF anymore.
Of course, there was no email about this, no notification -- nothing.
At a minimum, there needs to be better tooling for this -- it's just
too easy for something to fall through the cracks. But what I really
think we should do is simply end with this misguided policy. I
personally had no say in it. I didn't go to the FOSDEM developer
meeting. The first I heard about all this on this thread.
--
Peter Geoghegan
