> -----Original Message-----
> From: pgsql-advocacy-owner@postgresql.org
> [mailto:pgsql-advocacy-owner@postgresql.org]On Behalf Of
> greg@turnstep.com
> Sent: Thursday, March 20, 2003 10:24 AM
> To: pgsql-advocacy@postgresql.org
> Subject: Re: [pgsql-advocacy] PostgreSQL flamage on Slashdot
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
>
> > Jason, Greg,
>
> [snip quote from Jason]
>
> > But you see, that's exactly my point. We will *not* differentiate
> > ourselves from MySQL by slamming them at every opportunity.
>
> First, you should not address both of us but only respond to Jason's
> points. I in no way advocate slamming MySQL, but we should not ignore
> them either. Your initial argument was that we should not be competing
> against MySQL. The comparison between the two will continue to be
> made by many people for a long time to come. We cannot simply decide they
> are not worthy of our time and stand on the sidelines, waiting for the
> PostgreSQL vs. Oracle match.
Whoa, I never advocated 'slamming' either. In my 1st message I said I was
glad that we don't have to that. The word "hobby" is not nessasailry
slamming. There are lots of good "hobby" things, like Linux, that are great
products and leaders in their field. (Note: Linux is making the crossover
to "professional" now, please, no flames) (and in time, MySQL may make that
crossover too) I use 'hobby' vs. 'professional' to indicate robustness. I
don't know about you, but I'd not use a bank that used MySQL. If they used
Postgres, ok.
Most of the "battling" or "slamming" is done by the vocal uninformed few
that use incorrect arguments. "MySQL has transactions"+"MySQL is faster than
Postgres"!="MySQL is faster than Postgres when transactions are used", but
you'll find most people are arguing just that.
Furthermore, these same MySQL people don't bother tuning Postgres. We all
know how the config file is conservative. I think they are going to address
that in the next release: there have been looong threads about that already
in the PG-General list. In all my usage of the two, they seem to be
comparable in speed, YMMV. But I'd rather lose 10% maybe even 20% and have
all those great features.
How many people tune their disks with hdparm? How many people tune their NFS
block size? You can expect an even fewer amount to tune a much more complex
beast like Postgres.
-J
