Today is a good day to code

Oracle Buying Sun Microsystems

Posted: April 20th, 2009 | Author: | Filed under: Companies, Oracle, Sun Microsystems | Tags: , , , , | No Comments »

Today, we saw a game changer take shape.  Oracle’s bid for Sun Microsystems changes the landscape big-time for all technology players.  I personally do not love this situation.  I think that Oracle has a history of buying companies and scavenging them for IP and people, then discarding the rest.  I’ll go through who the biggest loser is in this deal, what will likely happen with the Sun’s major products, the biggest winners, and how it will impact developers.

The biggest loser in this deal is clearly IBM.  I believe now that IBM’s offer for Sun was a defensive deal.  They knew that if Oracle got it, IBM would become dependent upon Oracle to some degree in regards to the JVM.  This condition will force IBM to accelerate its work on a different JVM.  The problem here is that the IBM JVM will probably not be 100% compatible with the Sun JVM, creating further stress on the whole write-once-run-anywhere adage that Sun has been touting since the introduction of Java.  This will undermine the market for Java and send companies and developers to other alternatives to Java like Ruby and Python, which do largely the same thing and are still independent.

Oracle has in the past dismantled companies that it has purchased, distilling them down to strategic product lines, and profit centers and destroying and discarding the rest.  This is likely to be done with Sun.  What I see is the dismantling of Solaris, and the discarding of Solaris as a standalone product, and Oracle holding the IP.  Oracle will likely integrate the beneficial parts of Solaris into its Linux project.  JRuby will probably be killed, NetBeans will be killed, sadly.  Keep in mind when I say killed, these projects will continue, but they will need to be maintained and improved by the community, it is unlikely that Oracle will continue to expend company resources to work on these fringe projects.  Basically all of the minor open source projects that Sun was working on, will be subject to the same fate.  This includes the community version of MySQL.

MySQL will probably be split into two.  The MaxDB version of MySQL will gradually become Oracle 10g or whatever, and the community version will receive no further community updates from Sun / Oracle.  The community version of MySQL will lag very far behind the features of whatever Oracle version is its contemporary, and will likely stop being viable for most startups since the only upgrade path will likely be to an expensive Oracle license.

Why would Oracle do this?  For very basic reasons.  It removes a thorn from its side, in that they have been unable to capitalize on most web startups and even some of the larger web companies like Yahoo, Digg, and Facebook.  Now they will have the ability to extract larger licensing fees from these guys, and ( I’m sure they will believe ) mindshare from indie developers.  What really will happen is that these big guys will play along for a while, and then switch to PostgreSQL or something else.  I know that I am going to start looking more closely at PostgreSQL, and continue my heavy use of SQLite.

The biggest winners are, naturally Oracle, but perhaps less obviously, Microsoft and HP.  This is a huge opportunity to convince the cluster and server holdovers to embrace Microsoft Windows Server 2008 and HP server hardware.  In the confusion, some customers will move over, especially with some nice incentives to sweeten the pot.  Watch for Microsoft and HP’s sales departments to be a frenzy of activity over the next few weeks.  They will steal some business from Sun over this period.  Google could be a winner if they can commercialize their DalvikVM.  They are looking like geniuses right now for not building Android on top of the Sun JVM.

Developers will suffer significantly through this deal.  Sun had tons of developer friendly projects that didn’t make money, but gave developers a warm fuzzy feeling about Sun.  Small shops will rethink their dependence on the JVM and other Sun technologies.  Their developers will start recommending alternatives.  Not because anything has changed technically, but just because of their perception of Oracle as stodgy and difficult to work with.  Many of my favorite tools are Sun tools, but I wouldn’t necessarily pay for them as they don’t make my job that much easier; I will likely just fall back on my older tools that do the same things, but are just a little less efficient.

No matter the outcome, this is a dark day for tech, and it will take a while for the industry to absorb what just happened.  Sun was a community service, and it is difficult to imagine a world without them churning out cool Open Source projects.  I feel really bad for all of the MySQL AB guys who now work for for Oracle.  I would look for them to quit and create a new startup with a really cool database engine….