On managerial choosing (was: Postgres VS Oracle)

On Tue, Jun 19, 2007 at 11:22:17AM -0400, Chris Browne wrote:
> In cases where you *do not* spend the BPOM, there is not any
> particular evidence available to indicate that Oracle is, in any
> interesting way, more reliable than PostgreSQL.

I hate to say this, but as true as the above is, it has very close to
zero relevance to the way most senior managers make decisions.

It appears to suppose that the way decisions are usually made is
something like this: 1. Establish the problem; 2. Identify what is
needed to solve the problem; 3. Evaluate what available technologies
meet the requirements established in step 2.

The _actual_ way corporate decisions are made is mostly gut feel. The
simple truth is that most senior managers, even CIOs and CTOs, are
usually long past the period where technical detail is meaningful to
them.  They do not -- and probably should not -- know many of the
details of the problems they are nevertheless responsible for
solving.  Instead, they have to weigh costs and benefits, on the
basis of poor evidence and without enough time to get the proper
evidence.  Geeks who hang out here would probably be appalled at the
slapdash sort of evidence that undergirds large numbers of big
technical decisions.  But CIOs and CTOs aren't evaluating technology;
they're mitigating risk.

Once you understand that risk mitigation is practically the only job
they have, then buying Oracle in most cases is a no-brainer.  It has
the best reputation, and has all these features (some of which you
might not buy, but _could_ if your Oracle rep were to tell you it
would solve some problem you may or may not have) to protect you.
So, the only other calculation that should enter the picture is how
much money you have to spend, and how risky it would be to tie that
up in Oracle licenses.  In some cases, that turns out to be too risky,
and Postgres becomes a viable choice.  It's only exceptionally
visionary senior managers who operate in other ways.

There are two important consequences of this.  One is that competing
with MySQL is worth it, because MySQL is often regarded as the thing
one uses to "go cheap" when one can't afford Oracle.  Those people
will move from MySQL to Oracle as soon as practical, because their
DBAs often are appalled at the way MySQL works; they might get
addicted to the excellent features of PostgreSQL, though.  The second
is that marketing to management by using arguments, listing lots of
technical detail and features, and the like, will never work.
They'll ignore such cluttered and crowded brochures, because they
don't deal in technical detail.  We have to make PostgreSQL a
low-risk choice for them.

A

--
Andrew Sullivan  | ajs@crankycanuck.ca
The plural of anecdote is not data.
        --Roger Brinner