Red Hat to Acquire JBoss
Updated (11-Apr-2006, 17:00): Added reference to Stephen O'Grady's analysis up front, and news links at the end including a brief discussion on the apparent blog scrubbing at JBoss.
Red Hat has acquired JBoss. Here are the official links:
- Red Hat announcement
- JBoss FAQ (A good read.)
- Marc Fleury's blog entry on the subject.
- Stephen O'Grady's (Redmonk) excellent analysis is here.
At the very high level, this looks good. Two "open source software" companies, each using free software licenses, each pitching the enterprise and selling support and maintenance subscriptions. It's certainly a great data point for those that doubt an open source business model as the deal was valued at US$350M dollars.
I've never been a big fan though of thinking about "open source software" business models as something new or special, and think the more interesting questions are actually the boring business aspects of the deal.
These are actually very different companies:
- Red Hat is 11 years old, and has been public for 7 of them (this summer), with all the discipline to the numbers quarter on quarter that level of execution requires. JBoss is coming through it's explosive growth period over the past few years since emphasizing the support and maintenance model (rather than the training or consulting models of earlier JBoss Inc. history), and has a very different culture to support that growth.
- The Red Hat value proposition is inexpensive "UNIX" servers that scale out to add incremental compute power at marginal added cost, pitched to operations and procurement. The cost justification is easy against traditional big iron price points (i.e. it doesn't matter to the customer that the "license" is called a "support and maintenance subscription" or that this is open source in a direct way, but rather that it's much less expensive). The JBoss value proposition is inexpensive app server software, again that scales out to add incremental app serving power at marginal added cost, pitched to applications architects. The cost justification is easy against the traditional big app server licenses, and application architects no longer need to pervert their overall applications architecture to reduce the number of expensive app server licenses they need. Each company may be pitching the same enterprises, but their sales teams are probably talking to different people.
The processes and values of the two companies are likely very different at this point, so this is where the results of the acquisition will be interesting to watch over time. By values, I'm not talking about open source software, but rather all the boring business questions:
- What makes a good customer?
- Is this customer important?
- How are the field organizations compensated?
- How will next year's product offerings be determined?
If Red Hat can introduce JBoss technology to a lot of its customers, this could be great for business growth, assuming the packaging works. But the reverse may not be true: indeed Marc pointed out during the announcement of their Microsoft partnership last September that 50% of their user base could be on Microsoft Windows. Many customers may not want to move from their SuSE, Windows, Solaris, or other Linux systems to Red Hat's support and maintenance subscription.
This will be the interesting challenge for Red Hat — how to preserve the continued growth of JBoss technology regardless of the growth of Red Hat Advanced Server deployments. Novell is a JBoss partner. Microsoft is a JBoss partner. These could be opportunities for Red Hat to maintain the growth of JBoss, or to try to grow the RHAS base, and if they're too heavy-handed they run the risk of damaging JBoss growth. (I know they say they won't do this in the FAQ — let's see what decisions come out of their strategy meetings over the next few quarters rather than the intent of today's announcement.)
And this is the trick of any acquisition. If Oracle had acquired JBoss, it could have been an opportunity for Oracle to have grown its core business and revenue streams by making an inexpensive app server available as part of the mix, creating a better overall solution for their customers that need a database, apps and app serving infrastructure. This is similar to what SAP did a number of years ago with SAPDB (now MaxDB from MySQL). Oracle could have also used such an acquisition to begin the long and necessary cultural evolution to learn about open source as a supporting development, marketing, and delivery strategy, similar to Novell's long road with the Ximian and SuSE acquisitions. Or they could have made a complete mess of it, creating old-style pricing models, attempting to lock the technology to their way of doing things, and maintaining their 4-year old "open source" rhetoric about quality, security, and so forth.
This sort of cultural fit is important and Red Hat will make a much more predictable home for JBoss. One might wonder if Marc Fleury will become less outspoken than he has been in the past. I find that hard to believe for a couple of reasons. First, Red Hat already supports such a culture and has a number of outspoken people, including Micheal Tiemann and Ulrich Drepper. Second, he's always appeared to be sensitive to what information to share and what not to share. While I may disagree with a lot of what Fleury has to say with respect to "professional" open source and his perception of the battle with IBM, he's always remained quotable.
The Red Hat acquisition will hopefully prove to be a great fit and use of technology and the message to the benefit of all their customers, new and old, and to their continued growth.
- CNet coverage from Martin LaMonica (article on the acquisitoin quoting our own Dave Gynn, and blog in which the blog scrubbing is discussed).
- On the blog scrub issue (Register's article), I have to agree with Raven Zachary — who cares. Marc has been outspoken for ever, sometimes to the caustic and foolish point of the sort of material suddenly scrubbed from the JBoss blog site. Has he recanted? Probably not. Is he feeling perhaps a little sheepish or foolish — likely. Did ANYONE really believe someone wouldn't remember the comments and go cache diving? That's almost the more naive and embarassing thing at this point in history on the web.
- InformationWeek coverage from Charles Babcock
- ComputerWorld article from China Martens (and Eric Lai)
- CRN article from Paula Rooney