> I'm also pretty sure that they laid off their PostgreSQL support staff
> *before* the switchover to DB2; as you can imagine, they ran into some
> problems in the interval.
>
> Merlin, if you can actually provide a link, I'm sure that Tim P. would
be
> happy to give us a statement refuting IBM's interpretation.
My memory failed me. Here is the page I was thinking about (from his
famous 2 part article comparing mysql and pg):
http://www.phpbuilder.com/columns/tim20000705.php3?page=4
He never claimed that postgres was unstable, only that recovery was
nasty when it did go down (which was true in the 6.5 - 7.0 days). In
fact, he goes on to say that postgres was quite reliable. It could be
extracted from his writings that there were crashes, however. This
could be exploited in the usual nasty FUD way.
It would be nice to see some uptime statistics from him IMO. Not really
useful in a modern sense because it predates WAL, but it least to
contrast what IBM is talking about...
Merlin