[Discuss] Rob Conery's critique of MySQL?

[greg at freephile.com (Greg Rundlett (freephile))]

His critique pokes at defaults that MySQL ships with (default engine)
which he admits can easily be switched.  He shows how it gratuitously
substitutes zero for missing integer types when the schema says NOT
NULL, likewise for varchar where it will substitute the empty string
for a missing value.  He shows how if you change a field definition,
MySQL will truncate values (discarding your data) rather than warn you
that you can't change the definition without losing data.

He also does things like SELECT 1000/0 which MySQL reports as null.

I have no doubt that MySQL is and can be a terrific RDBMS.  However,
things have changed with the buyout of MySQL, then Sun and the forking
of the community.  A default installation of MySQL is dangerously too
flexible to be trusted with enterprise data.  PostgreSQL is not only
very fast, it also is a rigorous RDBMS that supports pretty much all
the things you can do with Oracle or SQL Server without the 6-figure
price tag.  The advancements being made to PostgreSQL as of late look
really interesting.

Greg Rundlett


On Sun, Jul 22, 2012 at 12:44 PM, Rich Braun <richb at pioneer.ci.net> wrote:
> For those of us lacking a 57-minute attention span to watch a full-length
> talk, what's the gist of Rob Conery's argument?  For many, I suppose
> PostgreSQL is a "default choice" but that isn't the case for most of the
> open-source tools I've used in the past 10 years.  I'm now on a project for
> which the default choice would have been Oracle, but the software architect
> has chosen MySQL as a cost-saving alternative.
>
> Why was this architect wrong?
>
> -rich
>
>
> _______________________________________________
> Discuss mailing list
> Discuss at blu.org
> http://lists.blu.org/mailman/listinfo/discuss