I've just been reading an article in PHP Architect magazine (http://www.phparch.com) which is the cover story for October called "Migrating from MySQL to PostgreSQL". I must say that this is a highly compelling article, especially for me, and is aimed at programmers that aren't necessarilly SQL experts or DBAs. For instance, like many PHP Web developers who use MySQL instead of flat files to store stuff! Instead of using a DB as a powerful tool. This article presents reasons as to why a more standards compliant DB is good for programmers, and why in some cases MySQL can be less of a friend to programmers than perhaps PostgreSQL.
I honestly believe that if the advocates of PostgreSQL wrote an article or case study along the lines of this article, it would go a long way to attracting many more programmers. In my experience all of the articles and tutorials are written from the perspective of why PG is a better DB as a DB. Rather than emphasise aspects like "PG is great because you can move complicated code like this <insert complicated PHP/Perl code here> ..... normally dealt programatically to your DB which can be both faster and applied to any other programmers (VB, Java) that you are sharing the important enterprise data with. I've not seen anything in articles aimed at PHP/MySQL users saying, "Hey, look at how these triggers can make your life soooo much easier" or "Hey, look at how cascading can save you oh so much coding" or "Hey look at all this programmatical logic that can be put into queries just by writing your own functions"
I have recently compared the PostgreSQL users to the Debian users (meant as a complement) by the fact that they are in general highly knowledgable of thier own subject and peripheral subjects too. They are passionate and well versed, and happy to nudge people in the direction of enlightenment without spoonfeeding them. But in the same way, the advocacy (IMHO) falls into the same boat as Debian. There is a certain self-assuredness that PostgreSQL is a far superior product and if someone can't see how obvious that is then maybe PG isn't for them (a little harsh I know but I'm trying to illustrate a point).
My point is that there are thousands, tens of thousands of programmers out there, that need to know why and how PG is so great. My eyes have now been fully opened by this article, and got rid of my nagging feeling that there was something great about PG that I "Just wasn't grasping, and couldn't put my finger on". Maybe the advocacy team should be aiming for all those programmers that desperately need PG, but don't know it yet, and probably don't have time to garner enough DB experience to understand why they need it!
Sadly the PHP Architect article is not free, I bought the electronic magazine for about $2, but believe it's worth every penny and more.
Just my 2 cents.
Apologies if the PG articles ARE out there and please notice that my comments do not say that they don't exist, but that I have never seen them.
Cheers
Tony.
---------------------------(end of broadcast)---------------------------
TIP 9: the planner will ignore your desire to choose an index scan if your
joining column's datatypes do not match
