At the risk of making it seem like this is all we’re talking about here at Pythian, here we go again.
Paul Vallee pointed me towards this article by John Dvorak that more or less echoes a blog post I wrote in French the day previous for my personal blog that you can read here: Le rachat de MySQL AB vu d’un paranoïaque; hopefully you read French and if not, maybe Google Translate will do a decent job; and if not well you can read Dvorak’s his is in English and he gets half the story!
So, first things first, Dvorak’s article leads me to think my theory is indeed wrong: if Dvorak says I’m right ; I’m afraid I’m wrong.
1. It’s just a shame because I’ve found new arguments to defend my theory :
You would agree the amount of money is suspect anyway. I don’t see MySQL enabling Sun to compete with Oracle, IBM, Microsoft or SAP on the software side. They’ve tried this for so many years. Becoming the new Redhat is just plain stupid. We don’t need another Redhat; It would have made much more sense for Redhat or Novell to do that (Except of course the price would have been very different). They’ve already spent millions to build OpenOffice.org, OpenSolaris and OpenJDK and what? Will they even make money from that? The fact they’ve bought 1 b$ a company that is worth 250 millions $ or less won’t guarantee anything like success especially as MySQL is more likely to kill Solaris rather than to help it to succeed.
Let’s do some parallels, if you use Google Docs we’ll agree the day the Internet will become really on the Air. OpenOffice.org will mean nothing compared to that Google thing. It will happen for the databases one day too and I’m not only talking about Amazon selling this kind of services, I’m talking about real TB databases sold as services with in-memory caching technologies on the website side to provide the performance you need. I’m talking about something you can rely on, highly secured. Google can work on that ! As they could work on any Open Source project to turn that into more reality than Sun with MySQL. And yes if Sun wants to make money from MySQL, they may want to build business cases where they’ll get money from Google and Yahoo! If they are annoyed, these Internet guys will stop supporting MySQL and go for another one; It’s just a database. (Funny that SAP just withdrew MaxDB, renamed again SAPDB, from MySQL a couple of month ago and will not support it for its ERP). On the other side Oracle is working to provide it’s database as a service and the CRM they’ve bought from Siebel has given them a real savoir-faire with SaaS (see this article for more details about selling database as a service) besides everybody knows more about SalesForce.com.
The “Will MySQL defeat the Oracle Database” question sounds to me like “Will OpenOffice.Org defeat Microsoft Office?” . The BEA deal just means they’ve lost in the middleware space and they’ve invented Java! if you look at what Oracle is doing on the Identity Management side coming from nowhere with OID and where Sun is still in the leaders, it’s very likely they will lose this war too. In addition, MySQL is no better than Postgres and even today there is still no guarantee MySQL won’t turn into another JBoss and will become as ubiquitous as “Linux” or “Apache”.
If you think I don’t know MySQL is better for purposes such as website or embedded technology, you’re wrong. What if Sun try to turn MySQL into a real transactional database ? They’ll just lose all its strengths Oracle bought as separate products from companies such as Timeten, Sleepycat or Tangosol. The once size fit all needs is over in here.
2. Now about Dvorak.
- His main target is to make audience which is not mine and he needs to say some radical things to get it
- He is ready to be wrong much of the time as a result and the most famous example was when he predicted the iPhone would be the biggest failure ever for Apple
- I like Fake Steve Jobs!
3, To be fair, I’ve listened again at the MySQL CEO and Sun Exec VP :
- I cannot imagine the MySQL CEO could possibly let such a thing happen
- I must say there has been previous episodes, the last one I remember was Sun announcing its Xen based product the same day as OracleVM, that are signs that Sun and Oracle don’t understand each other anymore. The Oracle support for Solaris x86 is the immediate second subject I can think of.
- They probably believe what they are saying even if they specifically avoid the questions regarding Oracle or SeeBeyond ; the fact Ray Lane was an exec when it happened, funny isn’t it ?
To conclude, I really don’t care who is in bed with whom, or who will win. Though with IBM buying Cognos, Ottawa may become a not so lovely place to live in the future.