Thoughts on 7.0.3 pg_dump addition

As everyone knows, there was discussion this week about a new utility to
be added to 7.0.3. The final decision was that the utility should go
into /contrib, and that the utility will not be mentioned in the release
notes.

I think we can learn some things from this episode.

Certainly, this had many missteps:

Company requests new feature with no discussion
Employed core member adds feature without discussion
Feature is added to a subrelease, which usually has no new features
Subrelease is already frozen
Feature is so minor, it is not even on the TODO list
Initially added as a new command that would disappear in 7.1, now
in /contrib

I realize there were extenuating circumstances that caused some of these
missteps.

The good news from all of this is how it was resolved. Each employed
core member had a different opinion, showing we still do control our own
opinions. I believe the final solution was optimal.

Having gone through this, I am now slightly less concerned about whether
the dynamics will change now that companies are involved with
PostgreSQL. Seems like this was handled just like before. In fact,
some employed core members even went beyond their normal inclinations to
maintain fairness.

We do have these goofy issues to resolve occasionally. The interesting
thing to me is that this company-induced one was really no different
than any of the past issues. We discussed it, and we resolved it.

Basically, it seems that while we will have missteps like this from time
to time, our ability to resolve them is unhampered.

Apologies to those who felt this week's discussion was too lengthy.
This is the only way we have to make fair, informed decisions that take
everyone's opinions into account.

--
Bruce Momjian | http://candle.pha.pa.us
pgman(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