Re: Do we really need a 7.4.22 release now? - Mailing list pgsql-hackers

From Greg Sabino Mullane
Subject Re: Do we really need a 7.4.22 release now?
Date
Msg-id 4c90db3061aa84d4646d11c9ab48b1c6@biglumber.com
Whole thread Raw
In response to Re: Do we really need a 7.4.22 release now?  (Gregory Stark <stark@enterprisedb.com>)
Responses Re: Do we really need a 7.4.22 release now?
List pgsql-hackers
-----BEGIN PGP SIGNED MESSAGE-----
Hash: RIPEMD160


> I don't understand this, as soon as we released 8.0 you could take that as
> advance warning that 7.4 was going to be desupported someday. So in that sense
> they've had four years warning that this time was coming. The fact that the
> date wasn't set in stone doesn't change their decision-making process.

That's silly, why would they think that? We did not have a policy when 8.0
came out as to how long branches were supported, nor was there any way
to gauge how often releases were coming out. Should people on 8.2 have been
thinking the same thing about 7 months ago when 8.3 came out? Recall that
our numbering has been historically somewhat arbitrary as well (7.4 should
probably have been 8.0, for example).

I guess I don't understand where Joe User was supposed to have gotten the
message that 7.4 was on its last legs. If anything, the fact that it is
on patchlevel 21 suggests otherwise. Us hackers and developers shudder
at seeing a 7.4 database, but there are plenty of businesses who are
still using it, and I think we owe it to them to give more advance
warning that no more patchlevels are coming along than 3 months.

Additionally, I don't know that it's possible to state up front how
long a particular major release is going to be supported, as some have
suggested. I think it's a great idea, but we tend to release "when
we are ready", so any guess is only ever a guess. Further, as our product
gets better and better, there will be an ever-increasing tendency
to *not* upgrade to the latest and greatest, since the version they
have already kicks ass, thank you very much[1].

[1] 8.3 rocks, by the way, thanks to all involved.

- --
Greg Sabino Mullane greg@turnstep.com
PGP Key: 0x14964AC8 200809191545
http://biglumber.com/x/web?pk=2529DF6AB8F79407E94445B4BC9B906714964AC8

-----BEGIN PGP SIGNATURE-----

iEYEAREDAAYFAkjUAv4ACgkQvJuQZxSWSsiWCgCeJTIAgYLpP1fAdEQ59Bmik22p
zGUAnRKciEyjpW9OqQey38JI24ei+bB8
=BRte
-----END PGP SIGNATURE-----




pgsql-hackers by date:

Previous
From: Andrew Chernow
Date:
Subject: Re: [PATCHES] libpq events patch (with sgml docs)
Next
From: Tom Lane
Date:
Subject: Re: [PATCHES] libpq events patch (with sgml docs)