Chris Browne wrote:
> I don't see any features that are in the release notes that seem like
> they strongly need to be added. The ones that seem plausible, but
> just a bit too marginal for the announcement, to me:
>
> - binary upgrade
>
I think that having in-place upgrade bundled with the database,
essentially blessing it as ready for serious use by the entire
development community, is the biggest improvement in PostgreSQL 9.0.
And, yes, I include the replication improvements in that ranking when I
say that. That's from both a technical (administrators will be happy)
and a FUD (remove that from the list of reasons not to use PostgreSQL)
perspective. Wander into any popular tech discussion forum when the
usual MySQL vs. PostgreSQL vs. everybody else discussions pop up, and
lack of binary upgrade is constantly mentioned as a reason not to use
the database. Saying "you can grab this migrator tool from this
site..." is a much better response than we had last year, but really
driving home that this is a solved problem for the database now is vital
for what I expect to be a widely distributed press release IMHO.
I submitted a response to Josh's survey which can get averaged along
with everyone else for the other features not on the list. I made my
other selections based on both the strength of the feature along with
how I perceived its ability to be understandable to those not already
using PostgreSQL too. So for me, VACUUM FULL rewrite is out just
because those not already using PostgreSQL know or cares what that
means, while things like the constraint improvements make a better
candidate because that's an understandable improvements to anyone
familiar with SQL databases.
--
Greg Smith 2ndQuadrant US Baltimore, MD
PostgreSQL Training, Services and Support
greg@2ndQuadrant.com www.2ndQuadrant.us
