Re: Call for 7.5 feature completion - Mailing list pgsql-hackers

On Mon, 2004-05-17 at 15:10, Bruce Momjian wrote:

> It is too late to think about pushing back another month.  We had this
> discussion already.  June 1 is it.
> 

I think I have to reiterate: PITR won't make 1 June. (I will be away
travelling soon). This has been said a number of times.

This is all rather disheartening, having laid out a time plan months
back, noting some of this. Yes, I am working on it, and no, I'm not hand
waving, but I do take time off every million keystrokes or so.

Mid to late June is a realistic time for the completion of PITR, given
the additional features I am now working on and the time allocation I
can reasonably give this (which in many ways is already unreasonable).
My last publicly stated completion estimate was right to within a few
days (mid-April, stated in early March).

I'm a very task focused person and I strongly support the notion of
deadlines and freezes. It's just in this instance, 1 June seems to have
been plucked from the air - unless there are other pressures that others
can see better than I. What are they again?

7.5dev is loads better than 7.4, though that was true in February.
Having waited until now, why not wait for the features? If somebody
suggested that we do 7.5 now and then 8.0 with Win32 & PITR in
September, I'd think about that, but I'm worried that the next major
release is a long way out after this next one, not a few months. Beta is
gonna take ages anyhow.

Many people in the community are still using 7.3 or below. If the
releases come too frequently, running an initdb is just a pain,
especially with out a "big reason" to sell to the business folks as to
why they have to put up with the downtime of upgrading (or time spent on
clever plans). "Is it running? Yeh. So why upgrade?" for existing users,
and "Does it have feature XYZ? No, but they think next release. OK,
well, we'll wait for that and then trial it" for new adopters.

Shipping too early is a bad thing too. It's not clear to me why you
would ship in a hurry when the community has waited years to get some of
the features on the URGENT list. Honestly, how long has PITR been
brewing? And, who thinks that we'll get increased adoption without the
big ticket items? (Even if your opinion is that PITR isn't one of them).

I can't complete by 1 June. Think worse of me if you choose.

Best Regards, Simon Riggs