Re: [HACKERS] What can we learn from MySQL? - Mailing list pgsql-advocacy

I have been thinking about this subject for a LONG time, and I hope I have
something to contribute.
>
> My question is, "What can we learn from MySQL?"  I don't know there is
> anything, but I think it makes sense to ask the question.
>
> Questions I have are:
>
>     o  Are we marketing ourselves properly?

I would say this is a clear 'NO!' When ever I read about open-source being
used anywhere, I always read MySQL. They are *very* good at this.


>     o  Are we focused enough on ease-of-use issues?

Again, NO! To often you guys settle for a work-around rather than a
feature. You are satisfied that symlinks will do the job. When someone
says they want a feature, you say, no - use a symlink.

Ease of use is VERY important, but few suggestions that address this are
ever really accepted. Yes, focusing on the functionality is the primary
concern, but "how" you set it up and deploy it is VERY important. You guys
need to remember, people are coming from a world where MySQL, Oracle, and
MSSQL all have nice setup programs.

I know a bit about this, as I made a "PostgreSQL for Windows" (It was
7.3.x) CD a while back. I had to do a lot of work on the postgresql
configuration, database initialization, and create a demo database. It
used a minimal cygwin environment, a Windows based installer, and some
custom function libraries. I tried to submit the configuration patch and
all I got was argument about using symlinks or how it wasn't needed.

The thing that kind of bugs me about this O/S project is that you guys are
a bit nit-picking about how someone uses it. I believe in the UNIX
phylosophy: capability not policy, flexability, etc. You guys seem to need
an absolute reason to include something, rather than a good reason to
exclude something. A lot of open source developers are turned off by this
sort of attitude.

>     o  How do we position ourselves against a database that some
>        say is "good enough" (MySQL), and another one that some
>        say is "too much"  (Oracle)

My argument against this is that MySQL is no less complicated than
PostgreSQL. PostgreSQL, in production is faster than MySQL, even though
MySQL may be marginally faster on some simple queries. The system resource
usage of both systems is very similar. PostgreSQL, however, boasts a lot
of standard features that make using it much easier.

>     o  Are our priorities too technically driven?

For the most part, you guys do a great, no .. fantastic, job at the
technical details of the database. Even though I get frustrated, I know it
is a great system. You *should* be technically driven.

If you want to blow the competition out of the water, you need a
non-forked Windows version of the database. You need a Java (or some other
portable environment) installer. You need to get out of the
hand-administered mentality of using symlinks and system level constructs.

One should be able to install the software, bring up a nice configuration
program which runs you through a few questions, and be done. This same
configuration program should be able to help maintain and control an the
installation. On Windows, have a service monitor program that starts and
stops the server, on UNIX, have it able to start/stop via init.d.
Everything else is "expert level."