On Sun, Feb 26, 2017 at 4:31 AM, Bruce Momjian <bruce@momjian.us> wrote:
> I think you are right. I was only guessing on a possible cause of
> Simon's reaction since I had the same reaction. When traveling, it is
> hard to get excited about reading a 100+ post thread that has reached a
> conclusion. I found Simon's summary of the 4 sub-features to be
> helpful.
OK, no problem. Basically, I think it's a bad plan to redesign this -
or add any large amount of incremental change to what's already been
done - at this point in the release cycle. Unless we're prepared to
rip it all back out, we've got to ship more or less what we have and
improve it later. I always viewed the mission of this patch as to set
the stage for future improvements in this area, not to solve all of
the problems by itself. I'm sorry if anyone was under a contrary
impression, and I'm also sorry that the discussion seems to have left
some people behind, but I did try my best.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company