Re: Forrester: Ingres and MySQL Lead Open Source Databases

On Tue, Aug 25, 2009 at 12:34:40AM -0700, Ron Mayer wrote:
> Greg Stark wrote:
> > On Sun, Aug 23, 2009 at 8:21 PM, Josh Berkus<josh@agliodbs.com> wrote:
> >> So in the case of analysts, it's a case of "can we influence this
> >> analyst to produce a favorable report?  How much will it cost,
> >> and how many people will it reach?"  Not an argument of "are
> >> analysts good or bad."
>
> Perhaps also add the question "what's the most cost effective way to
> influence the analyst?" as well.
>
> > Well, how much would it cost? Can you outbid Oracle, Microsoft,
> > and IBM?
>
> ISTM there are exactly 2 ways once can effectively influence such
> analysts to say that Postgres is better than the alternatives.
>
> 1. For ethical analysts, the most cost effective way - and
> practically the only way - is to produce a better product than the
> alternatives.

I wish that were enough.  This type of analyst type needs to have a
point of contact and that point of contact has to have backups just in
case, which can be handed off to the aforementioned analyst.
Maintaining a point of contact for such analysts is not cost-free, and
Josh Berkus has been doing an admirable job of being that person.  He
could really use some backup people.

> 2. For less ethical analysts, the most cost effective way is
> probably to spend money on them - *and* do the legwork for getting
> favoriable data for the report.  I don't doubt that if someone
> wanted to buy a report from Forrester to address the question "can
> postgres scale to handle databases like skype's" and then gave them
> willing skype contacts as references, they could write a glowing
> report.   But does that really accomplish much?   It gets one nice
> report, but has a one time effect while any resources spend on #1
> has recurring effects until the competition catches up again.

Bribing a Forrester analyst is not really in scope for a FLOSS project
of any description.  Making sure that their emails, phone calls, etc.
get returned promptly and in a coordinated fashion is, and we can do
that.  It takes efforts of a different kind from database kernel
hacking, and fortunately, there are people who believe this is worth
the time.

> I'm guessing that the most cost effective way to influence analysts
> in the long run is to spend the resources making sure the product is
> better than the competition.

I wish it were this way.  We already have a product that's better than
the competition.  We have for many years.  What we now need is to
further expand our efforts to let people know.

Cheers,
David.
--
David Fetter <david@fetter.org> http://fetter.org/
Phone: +1 415 235 3778  AIM: dfetter666  Yahoo!: dfetter
Skype: davidfetter      XMPP: david.fetter@gmail.com

Remember to vote!
Consider donating to Postgres: http://www.postgresql.org/about/donate