Greetings,
* Jonathan S. Katz (jkatz@postgresql.org) wrote:
> On 9/11/18 10:08 AM, Peter Eisentraut wrote:
> > So that would mean that if the EOL month is September and the next minor
> > release is scheduled for November, but we do an unplanned release in
> > October, we would then have to still support it until November? Or if
> > we skip the November release, we have to keep going until February?
> >
> > I think a bit of ambiguity is good here. After the EOL month, you're on
> > your own. We might do something afterwards for technical or bookkeeping
> > reasons or because we think it's important, but don't count on it.
> > Maybe in five years we'll be releasing a minor every three weeks, how do
> > we adjust the policy then? Let's not over-specify this.
>
> +1. I think the first sentence in the "PostgreSQL Release Support Policy
> section clearly states what happens:
>
> "The PostgreSQL Global Development Group aims to support a major release
> for five years. After its end-of-life (EOL) month ends, a major version
> receives one final minor release. After that final minor release, bug
> fixing ceases for that major version."
Yes, that's what it says, but I don't agree that it's clear at all.
That's exactly what started this thread.
> That said, maybe it's better to make that the first sentence on the
> page, as really that's the most important part. Details on how things
> are numbered, upgrades, etc. could go closer to the table.
A big block of text above a table which contains the actual information
people are looking for is just going to get ignored until someone
explicitly points out that "well, this is what the policy says".. and
then further explains it, because it's really rather confusing. We all
get it and understand it because we've been around it for a very long
time and are familiar with it.
Thanks!
Stephen