Re: Pluggable Indexes (was Re: rmgr hooks (v2))

On Wed, 2009-01-21 at 19:13 +0200, Heikki Linnakangas wrote:

> > You don't want pluggable indexes, don't use 'em. But that isn't an
> > argument against allowing the capability for others. That line of
> > thought would have led us to banning pluggable languages. We should
> > respect the roots of this project and look for ways to enable the
> > advancement of database technology, not limit it to only how far we
> can > currently see ahead through the fog.
>
> This is an open source project.

That's a whole different discussion.

Extensibility is what gives options in production. Yes, the academics
can do whatever they like. We know the reality is people don't fiddle
with core code for a range of reasons but are happy to use extensions.
I'm in favour of allowing people that use Postgres to get access to
advanced technology without asking my permission or paying me a licence
fee for a modified version.

We support extensible everything, but not indexes. Why?

PostgreSQL is supposed to be The World's Most Advanced Open Source
Database. There is no good technical reason to hold back this patch.

The arguments against this patch seem to revolve around fears of
commercial exploitation or subverting the release process. Or telling
people that we know better than them and they can't possibly write an
index worthy of actual use. They might not be able to, its true, but I
see no reason to prevent them either.

> *That's* how this project moves forward.

We've got one committer working almost exclusively on new indexes.
Preventing work on new indexes by non-committers has meant that Bitmap
indexes, which first came out in 2005 have not been usable with
Postgres. That forced people *away* from Postgres towards Bizgres. Lack
of Bitmap indexes is a huge issue for many people. It's 2009 now and it
seems probable that without this patch it will be 2010 at least before
they see BMIs, and later still before they see other index types.

Many people can see the blockage there. I agree it is right to have
prevented BMIs from being committed to core, but they have been usable
and beneficial for many years now for read only workloads. In the
current way of thinking early GIST would never have been allowed in and
there would be no PostGIS.

--
Simon Riggs www.2ndQuadrant.com
PostgreSQL Training, Services and Support