There was a time when vendors knew how to color inside the lines. A database vendor sold databases. An operating system vendor peddled operating systems. And application server vendors were in the business of selling application servers.
Customers knew what "application server" meant, which is what paved the way for low-cost, high-value open-source application servers like JBoss, Geronimo, and others to arise. The category was well understood. The only thing the customer had to decide was whether she wished to overspend on a brand-name application server or buy into an open-source upstart.
As the economy continues to pressure IT budgets, a new breed of application server is rising, one that doesn't color nicely within the lines of the traditional "app server" definition.
First there was SpringSource, which in April 2008 claimed it had developed the first "proper Java application server" to hit the market in more than 10 years. IBM, Oracle, and Red Hat (JBoss) must have found this surprising, given that "proper Java application servers" were what they presumed to be selling to their customers.
SpringSource, pressing the issue further, this week announced Spring Roo, an "interactive, lightweight, user-customizable tooling that enables rapid delivery of high-performance enterprise Java applications." At just 4MB, Spring Roo is tiny, but its promise is big: "working applications within 10 minutes of finishing the download" (video here).
Red Hat is taking notice of SpringSource's moves, but SpringSource founder Rod Johnson is dismissive of Red Hat's ability to compete:
Red Hat recently announced a defensive move (JBoss Open Choice strategy) motivated by trying to play catch-up with SpringSource. Clearly the momentum of SpringSource tc Server and dm Server has Red Hat worried, along with the continued advance of the Spring Framework as the de facto standard component model for enterprise Java.
The "JBoss Open Choice strategy" appears to be a repackaging, rather than new technology, which attempts to position JBoss as still relevant in a brave new world of changing requirements. On a positive note, it appears Red Hat has finally realized that many developers and customers have long since moved away from the full Java EE stack; that the traditional heavyweight application server has declined in importance; and that the Spring programming model is important to their customer base.
The paint is barely dry on JBoss' incursion into IBM's and Oracle's application server territory, and already it's under attack from SpringSource, which is positioning JBoss as old technology, despite barely being out of its teens. It's a sharp attack, but perhaps a sign of open-source competition to come.
Consider MindTouch. The entire idea of a Java application server got a shaking this week from MindTouch, which announced a new application packaging feature for its collaboration platform, as TechCrunch reports. This feature:
(A)llows developers to create a compressed file for import into other MindTouch instances, letting enterprise users install add-on applications easily. This addition represents MindTouch's ambitions to become an application platform where installing applications (is) as easy as adding Firefox add-ons.
This may not sound very groundbreaking, until you consider that MindTouch is essentially announcing that it has turned the wiki into an application platform. Until last year MindTouch was mostly known for its DekiWiki technology. This application-packaging technology basically allows customers to build on wiki collaboration, which is much more than just a MediaWiki-style wiki, as ReadWriteWeb notes.
Neither SpringSource nor MindTouch fit the old application server mold, but it's not clear customers should care. Enterprises just want to get work done.
Just as Google is shaking up e-mail with Wave and Wolfram Alpha is redefining a corner of search, so too are SpringSource and MindTouch redefining what application server means for the modern enterprise, while open-source companies like Acquia redefine Web content management, Marketcetera pushes the envelope on financial trading platforms, etc.
Open source is the new innovator, and not merely the commodifying force in the market.
Indeed, with traditional software markets perhaps reaching a critical saturation point, we may see much more "coloring outside the lines" in enterprise software from open-source projects and companies. Oracle's response has been to consolidate, but others seem to be responding with a different strategy: innovation.
Disclosure: I am an adviser to MindTouch. Red Hat is a business partner. Acquia is in some ways a competitor.
Follow me on Twitter @mjasay.
I didn't know all the history behind JBoss until I read Shaun Connolly's blog on the topic. It turns out that JBoss has been stalking BEA for some time. Unfortunately for BEA, IBM and JBoss now appear to be poised to eclipse its once-thriving business altogether.
JBoss never should have happened, as Bob Pasker once wrote. BEA ceded the field of battle to JBoss by fixating on IBM:
... Read more
Credit Suisse is citing Oracle's acquisition of BEA Systems as a "win win across the board" in research sent out to its clients. Writes Jason Maynard:
It further diversifies Oracle's business and strengthens its competitive position....We continue to think Oracle is the best positioned enterprise software company and the most capable of handling a potential slowdown in IT spending.
Rebecca Wettemann, VP of Nucleus Research, piles on to the lovefest, arguing:
Acquiring BEA technology cements Oracle's position of leadership in the application and application integration space. While Oracle's integration story with Fusion middleware and AIA was already much stronger and "realer" than SAP's, this adds more integration expertise and technology to the portfolio.
So, good for Oracle. But it's also probably a huge boon for Red Hat and the JBoss team. Why?
... Read moreApparently, Larry Ellison didn't like the idea of anyone getting more than a few minutes of press for an acquisition. Following on the heels of Sun's acquisition of MySQL, Oracle is announcing that it is buying everyone else on the planet to prevent this from happening again.
No, actually, Oracle has announced that it is buying BEA Systems for $8.5 billion. The proprietary world just got even more consolidated with less choice and higher costs for buyers.
... Read morePaul Kedrosky paints a not-so-pretty picture of BEA's future. Now that it has rejected Oracle's offer of $17/share through pocket veto, all the while begging for a higher price, it's almost certainly going to get a lower offer the next time around.
What happens next is rarely pleasant. Absent a change of heart, or a new offer somehow emerging, the initial acquirer will likely come back and make another offer -- this time, however, for less than was originally offered. Why not? After all, the market just told it that it offered too much, so it might as well offer less.
That is what the future likely holds for BEA. It will almost certainly be bought, but likely at a price under the price Oracle originally offered for the company. Ouch.
BEA took the high road. Now it's likely to get run over.
BEA Systems yesterday entered the final phase of capitulation to Oracle, throwing out a "but we're worth so much more than $17 per share!" counter to Oracle's offer.
Most acquiring companies would take mercy on the desperate gesture of BEA at this point, but Oracle might very well hold out.
Oracle is the sort of company to drag the companies it woos through a bit of hell and humiliation to make them good and desperate to be acquired at its price. This is very likely what will happen with BEA.
In other words, BEA's plea for more money will probably result in the opposite: an even lower offer a year from now, after BEA's public subjugation to Oracle is complete. Oracle is demanding an answer by this Sunday, but it can afford to wait much longer on the response it wants.
In the meantime, BEA will flounder, just as PeopleSoft did before it.
... Read moreWhat does JBoss gain from it? Potentially a lot, at least in the short term.
So who gains? Well obviously Icahn and co and its 12% stake in BEAS are pocketing the full 30% acquisition premium, which is a cool $200m. Not bad, take the money. The shareholders gain, Oracle gains. Ironically IBM and RedHat will see positive windfall from this move in the short term. Customers that were on the fence with BEA renewals and don't like the idea of Oracle controlling too much of their account are going to be persuaded to move. ... Read more
I'm loving doing this podcast series with The Register's Ashlee Vance and MuleSource's Dave Rosenberg. We dubbed it "Open Season," and often with good reason. In this last installment, we talked about a range of open source topics, including:
- Patent protection and open source, focusing on Novell's Moonlight project
- Ubuntu's mini server OS for VMware
- The 'Calista Flockhart' hypervisor from VMware
- Net App sues Sun and Fishworks
- BEA disses open source
- InfoWorld's Best of Open Source list
- Developers and salespeople learning how to do open source
- Good documentation equals dollars
- Other topics....
It's not poetry, but it's fun and occasionally insightful.
- prev
- 1
- next
