How can we help? - Mailing list pgsql-hackers

Hi all!

Sorry for the length of this but I'm trying to get an idea of where my
company can contribute to the best effect so I have a number of questions.

To begin with my team and I have some energy/time/$ over the coming months
to put directly into PostgreSQL-related development work. Technical Pursuit,
has made a commitment to PostgreSQL for both our internal projects and our
customer projects. We've been evangelizing PostgreSQL for a while now (I did
a talk on it at the Database Discovery cruise last June in Alaska -- a lone
voice in literally a sea of Oracle folks) and have started doing
Oracle-to-PostgreSQL conversions for customers wishing to transition away
from Oracle. We're also getting ready to ship a beta release of our TIBET
product that uses PG as the backend source code repository among other
things.

Areas we have customer/business needs in include replication,
backup/recovery, monitoring/control, XML support, HTTP/HTTPS protocol
support for postmaster, pl/pgperl, possible pl/jython, and possible
compile-time inclusion/configuration of time-travel (--with-time-travel ?).

On the process side, is there an IRC or other chat-based system in place for
the PG team to coordinate their efforts? If not, would an IRC system hosted
by TPI be something folks would be interested in using? We'd be willing to
start hosting a set of IRC channels if that would assist the team and the
community in support issues etc.

For XML support I've contacted John Gray who did the current XML contrib but
has since ceased development and he's granted me permission to pick up where
he left off to improve XML support as it relates to his contrib module. Is
there any move underway to integrate XML with PG at any other level?

If we were to contribute to replication/backup/recovery solutions who would
we coordinate that effort with? If that's something the core team is on top
of, can we get an idea of what is expected by August so we can advise our
customers and plan accordingly?

What is the planned status of Java support in the engine? Is there anyone
working on JVM integration at this stage and if not, how could we best
integrate with the team to take on this task? We're looking seriously at the
idea of a pl/jython that would leverage the Jython language to provide
scriptable Java procedures in the engine. Any feedback on that idea?

For several scenarios we see value in having the postmaster respond to
http(s) protocols and relay to internal functions for processing of web
traffic. We've got a simple perl "bridge" that performs this task now and
the performance is more than adequate but we're sure it would be even better
if we could avoid having the separate perl component. Is there any interest
in this elsewhere? Any feedback on how/where we should start other than
hacking postmaster? ;)

What is the current thinking on re-introducing time-travel support, perhaps
as a compile-time feature ala --with-time-travel? This is a feature our
customers would get significant financial benefit from. I've seen recent
notes that this might be possible to do. We're *extremely* interested in
this direction given the potential for differentiation from other products
in the marketplace. It would strengthen PG significantly in our mind.

Finally, we're starting to do a lot of work in pl/pgperl(u) and are
wondering whether that's an area we can again contribute in. If so, how do
we get involved? PG/Financials anyone?

Again, sorry for the length of this and the raft of questions. I hope you
understand we're in a somewhat interesting position with some time and $ to
focus on PG, particularly as it relates to replication/recovery issues. Any
guidance on how we can put that to work for the community would be
appreciated.

Thanks.

ss


Scott Shattuck
Technical Pursuit Inc.