Re: [HACKERS] A small problem with the new inet and cidr types

I am replying to something posted by D'Arcy on November 2nd, so I will
quote the whole thing:

> Thus spake Tom Ivar Helbekkmo
> > While we're talking of release cycles, I'd like to take the liberty to
> > post something that I've been keeping around for a while, which was
> > written by Dave Burgess (of *BSD FAQ fame) some time back. He was
> > posting to a NetBSD mailing list, so there is wording in here that's
> > specific to that environment, but there is much good, general truth
> > here. With respect to the CIDR case, note especially his "get it
> > right or get it out" rule. Translated to PostgreSQL 6.4 terms, we
> > get: "Whoah! This wasn't what we wanted! Decide quick: do we ship
> > what we have, or do we ditch it and do it right for 6.5?". We chose
> > to do neither, and there's a lesson to be learned from this.
>
> Perhaps but let me point out a few things.
>
> - The code wasn't that late. Other things were holding up release
> and the inet/cidr stuff was in by the drop dead date we were given
> at the time.
>
> - We may not have followed the release method you quote but that
> wasn't our release method at the time. Perhaps it should be
> but let's not be blinded by hindsight.
>
> - The latest problem has turned out to be a general problem with
> the system. Rather than being an added bug, having this in now
> just allowed us to find an existing problem sooner than we might
> have otherwise.
>
> - Because of a few late night coding sessions we now have a really
> cool feature and we beat the market. That's something we can use in
> our announcement. Putting this feature in, even if we pushed the
> deadline a little, didn't weaken the product, it strengthened it.
>
> I don't want to sound defensive but people have been sounding negative
> about having this type in. I just want to make sure that we all see
> the positive side of this too.
>
> > Let me also take this opportunity to submit a plug for "The Mythical
> > Man-Month" by Frederick P. Brooks. Anyone involved in software
> > development who hasn't yet read it, should. Get the 1995 special
> > edition, expanded and edited from the original.
>
> I'll second that.

OK. I want to take responsibility for the one-month delay of 6.4.

If people remember, I posted the mega-patch mid-August, which added
multi-key indexes to the system tables, improved the heap API with
proper locking, and improved system cache usage.

This patch generated several problems, some of them bugs, some of them
areas I missed in adding multi-key system indexes. We couldn't go live
with beta on September 1 because we had major problems doing initdb on
some platforms. We kind of started the beta, but without all platforms
participating, it was pretty useless, and we started to let features in
during the month while we struggled to get the mega-patch problems
resolved.

We didn't start serious beta until October 1, and D'Arcy got caught with
Paul Vixie defining/adding the IP type. D'Arcy was just left holding
the bag when people started to worry about the delay. It was not his
fault.

It was me, putting a 18k line patch in two weeks before beta. Of
course, I started earlier than mid-August, but it takes me a while to
generate 18k lines of diff. :-)

Just wanted to clear that up.

--
Bruce Momjian | http://www.op.net/~candle
maillist(at)candle(dot)pha(dot)pa(dot)us | (610) 853-3000
+ If your life is a hard drive, | 830 Blythe Avenue
+ Christ can be your backup. | Drexel Hill, Pennsylvania 19026