More to the point though, I think this is a feature that really really
should be in the DB, because then it's trivial for people to use.
How does putting it into PGPool make it any less trivial for people to
use?
The answers are at
http://www2b.biglobe.ne.jp/~caco/pgpool/index-e.html . Specifically, it's a separate application that needs configuration, the homepage has no real discussion of the potential pitfalls of pooling and what this implementation does to get around them, you get the idea. I'm sure it's great software, but it doesn't come as part of the DB server, so 95% of people who would benefit from query caching being implemented in it never will. If it shipped with and was turned on by default in SUSE or RedHat that would be a different matter. Which I realise makes me look like one of those people who doesn't appreciate code unless it's 'popular', but I hope I'm not *that* bad...
Oh OK, I'll say it, this is a perfect example of why My*** has so much more mindshare. It's not better, but it sure makes the average Joe _feel_ better. Sorry, I've got my corporate hat on today, I'm sure I'll feel a little less cynical tomorrow.
M