Re: what's going on with lapwing?

On Thu, Mar 6, 2025 at 2:13 PM Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> It's hard to "mandate" anything in a distributed project like this.
> I don't really see a need to either, at least for cases where an
> old animal isn't causing us extra work.

I don't know, to me it feels like we have the argument about whether
StegosaurOS is actually dead or whether there might be survivors of
the Chixulub impact hiding somewhere several times a year. The
participants in those discussions are always pretty much always the
same, and their opinions are pretty much always the same, and the
threads are long and tiresome. Eventually we usually get consensus to
forcibly retire some OS that's been EOL for 5 or 10 years and then we
do it all over again the next time somebody's losing their mind. I
think all of that energy could be better spent.

What really makes this unfortunate is that we typically only have this
discussion after that machine has ALREADY caused a bunch of hassle for
a bunch of people. You commit something, the BF turns red, you fix it.
You do the same thing again. Then maybe a third time. By the time we
actually get consensus to remove things, they're generally not just
MOSTLY dead. They're at the point where all you can do is check their
pockets for loose change. They have kicked the bucket, shuffled off
this mortal coil, run down the curtain and gone to join the choir
invisible. They are ex-operating systems.

It just doesn't seem reasonable to me that you have to basically show
up and prove that you've already wasted 100 hours or whatever on a
defunct OS before we'll consider giving it the boot. It's a
predictable outcome. Once the OS is dead, you can't easily upgrade to
newer software versions, which is eventually going to break something
for somebody, not to mention that eventually no PG developer other
than the BF member owner will be able to get access to a copy of that
OS to fix anything that breaks. It's only a matter of time before that
becomes an inconvenience.

--
Robert Haas
EDB: http://www.enterprisedb.com