On Mon, Oct 03, 2005 at 12:45:05PM -0700, Josh Berkus wrote:
> Jim,
>
> > If we're going to go with 2pc, it should be 2PC IMO. But I think it
> > would be better to spell it out (two-phase ocmmit).
>
> Oh, right, missed that.
>
> > New Advanced Features, para 2:
> > s#INOUT#IN/OUT#
>
> Really? I thought the technical term for the 3 types of params were IN,
> OUT, and INOUT.
Well, it's inconsistent with the usage in the first paragraph (IN/OUT).
And usually when I see a reference to the *concept* (which is what I'd
say the PR is), it's usually called IN/OUT as an umbrella term for
IN/OUT/INOUT.
> > The table partitioning section is a bit misleading in that it doesn't
> > actually help with large tables, it just makes partitioning a bit more
> > effective.
>
> Suggested replacement language?
Got a meeting so this is just a quick hack...
...the query planner is now able to entirely avoid scanning tables that
it knows will not have any data to satisfy a query. This feature, called
Constraint Elimination, utilizes constraints placed on tables to decide
if they have any valid data for a given query. This is the foundation
upon which table partitioning can be built.
> > On the BSD/GPL horse, what about...
> >
> > PostgreSQL is distributed under a BSD license, which unlike GPL licensed
> > software, allows use and distribution without fees for both commercial
> > and non-commercial applications.
>
> I am *NOT* picking a fight with the FSF. Also, many people (myself
> included) do not agree with MySQL's draconian interpretation of the GPL.
>
> This would get us headlines, all right, but the wrong kind.
Is there such a thing? :)
In any case, I definately think we need to raise awareness of the
licensing issue.
--
Jim C. Nasby, Sr. Engineering Consultant jnasby@pervasive.com
Pervasive Software http://pervasive.com work: 512-231-6117
vcard: http://jim.nasby.net/pervasive.vcf cell: 512-569-9461
