So Monty now claims that Oracle is MySQL's main competitor. Given that for years MySQL AB has said the exact opposite is kinda odd. They did have a short run in with SAP but that didn't go any where .. well it resulted in MySQL loosing focus, adding tons of barely functioning features which ended up in this entire roadmap debacle we are still suffering through. Anyways the EC has thrown in their hat to stop the deal. Björn seems to share their concerns and so do many others it seems. I still don't get it. This is my attempt at clarifying my point of view and also making sure that people remember the proper terminology (stop using commercial and proprietary as synonyms).

Ok, seems like this is one of my yearly rituals: Trying to build some momentum behind PDO. So far it has remained lingering after every such attempt, not that this isn't partially at least my fault. I think last time I said I would work on the tests and in the end I didn't get anywhere. Unfortunately nobody came around to kick me for this lack of commitment and so the idea faded again. So enough with the gloom today is a fresh start and I think one thing has changed significantly since then: We have people posting patches for PDO on internals and someone (aka Matteo) that has been fairly steadily trying to close some tickets. Pierre also seems to be interested in putting in some time. Scott has always been committed to the SQLite driver and Ilia and Felipe have also been good for quite a few bug fixes in the past. So it seems we are closer than in quite some time to have the critical mass of people that care to get things going again!

So I guess the argument that Monty is putting forth is that MySQL's future depends on some company exploiting their ownership of the MySQL copyright. As in, he feels that MySQL future will fall apart if there isn't a company that makes its money by dual licensing MySQL under some proprietary license. Amazingly enough Stallman, chief hater of all things proprietary, agrees. Hell seems to have frozen over. I however humbly disagree. The MySQL community has already left the dual licensing model behind. Sun has never shipped the "best" MySQL version, the community has!

I am currently touring the US as part of the CodeWorks 09 conferences, where I am holding a 3 hours tutorial on database driven PHP development. So far my sessions have consistently been less attended than the talks on testing, OO and security that are running as well. So it goes. I guess the plus side for the attendees is that I can spend more time addressing their personal needs. My tutorial essentially is based on all my previous talks on database for which I have updated all the relevant slide sets. I have also added an introduction to PDO to get things started. Since all of that will never fit into 3 hours, I let the audience pick which talks they want to hear. Unfortunately so far the Doctrine ORM talk hasn't been picked, but I am getting better and better at holding the talk about the telephone directory application and I think the work I put into making this talk very much code based is really playing off. However the slide that seems to get the most "aha, that is really cool!" is the slide about optimistic locking.

Long ago I had a project where we needed to generate Word documents. We ended up using a fake X-Server running OpenOffice and some queuing mechanism to manage the fact that each OpenOffice instance could only handle one conversion job at a time. No fun, but it worked, does not scale amazingly obviously. Back then I of course also tried the KOffice converter which is easily run with little overhead via a CLI interface. Unfortunately it wasn't really up to the job in terms of Word compatibility.