Oracle to buy JBoss (and others)

tfritsch writes "According to a story at News.com it looks like Oracle's shopping spree is to continue. The JBoss acquisition could be big - what does it mean for the future of the JBoss Application Server?" From the article: "Oracle makes the majority of its revenue from its database and applications business. And it has its own line of Java middleware, which competes with JBoss' software, and a set of Java developer tools. However, Oracle has been warming up to open-source products, including Zend's PHP development tools, over the past year because its corporate customers are increasingly using open source software, according to company executives. "

Comments

Don't trust Oracle

[AKAImBatman]

Don't believe for a minute that Oracle would purchase JBoss to "help it shift customers to a subscriber-based model". Oracle already has a superior J2EE server based on Orion technology. Far more likely is that Oracle wants to pull another PeopleSoft aquisition. They'll buy up JBoss, kill the company, then let the product die on the vine. All while pushing how "Open Source Friendly" they've become.

Re:Don't trust Oracle

[Anonymous Coward]

JBoss is the best performing JVM out there, bar none

What an amazing load of crap. JBoss is a JVM? And here I thought is was an EJB container cum J2EE server.

Mods, how about blasting this idiot back to the stoneage?

Re:Don't trust Oracle

[AKAImBatman]

Killing off the leading product in a high growth market is bad business. It doesn't appear logical.

1. JBoss is not the leading product in the J2EE market. It's a competitor, but nowhere near the top.

2. J2EE servers are not a high growth market. In fact, the market is oversaturated at this point, with servers from BEA, Sun, IBM, Novell, JBoss, Apache, Macromedia, ObjectWeb, Pramati, Borland, Orion, Oracle, Caucho, Apple (!), ATG, Compaq/HP, Fujitsu, Gemstone, Hitachi, IONA, Secant, Sybase, and quite a few others who aren't worth mentioning. Of those, Apache and ObjectWeb directly compete with JBoss to provide an open source J2EE server. Nearly the entire market competes with JBoss for support contracts.

Re:Don't trust Oracle

[AKAImBatman]

JBoss is the leading open source java app server

That's a of qualifiers. The question is: Do they mean anything?

I can unequivolcally state that I am the leading Slashdot poster with Batman in my name. That statement doesn't generate revenue or otherwise help me in any useful way.

OSS Java app servers with low/null acquisition costs are a high growth market.

According to who? I have observed no real push by the market to move from their expensive servers to OSS servers. There is a push for cheaper servers like JRun and Caucho, but JBoss doesn't really enter into play. Many companies eschew a J2EE server altogether and use just a servlet container like Tomcat.

That big list you mentioned has how many OSS competitors for JBoss in it?

Two direct competitors. There's also noncertified servers such as Exolab, OpenEJB, Enhydra and non-EJB servers such as Tomcat, Jetty, OpenJSA, Gefion LiteWebServer, and many projects/companies that went defunct shortly after producing viable competitors. Why did competitors like Exoffice go MIA? Well, it seems the market isn't big enough to support that many competitors.

OSS apps with comparable feature sets tend to grow their userbases at the expense of commercial applications. It is hard to compete with free.

It is hard to compete with free. Which is why JBoss can't compete with SunOne, HP-AS, Orion, and SybaseAS, all of which are either outright free or provide free editions bundled with other services. Which would a large company rather go with: A free OSS product or a free commercial product? Most go for the latter as they feel that it guarantees stability, support, and upgrades.

Unless you've got some hard numbers to back up your assertions, I'm afraid that the massive, OSS, J2EE market simply doesn't exist.

Re:Not gonna happen.

[AKAImBatman]

Larry can't just write a check, get regulatory approval and be done. There is no way Marc will sell his baby...

You might be amazed at how much power is contained within a single zero. Throw enough of them on the check, and even Marc would have a hard time resisting.

A discussion on the PostgreSQL advocacy list...

[tcopeland]

....right here had an interesting comment from Bruce Momjian:

It is interesting that they are purchasing companies that almost fully control the software but give it away free as open source: Sleepycat, JBoss, and Zend. Oracle's purchase months ago of InnoDB used by MySQL was a similar move. What they are _not_ getting involved in is software that is community controlled, like PostgreSQL or Linux, because it much harder to see how a purchase would allow tight control of the software, resulting in revenue.

Rather well said.

I've been pleased with Oracle's JDeveloper; writing an extension for it has been interesting and the Oracle folks have been quite helpful.

OSS projects selling out?

[Caspian]

I've long operated under the assumption that any decent (or even semi-decent) piece of "freeware" (free as in beer, but not as in speech) for Windows will eventually sell out and become "shareware" and/or conventional commercial software. Likewise, I've assumed that any decent piece of "shareware" will slowly go the route of full commercialization. This assumption has served me fairly well. (Examples of this pattern: PowerArchiver used to be freeware; now it's shareware. Paint Shop Pro used to be shareware; now, it's being sold in stores.)

Am I now going to have to start assuming that any decent OSS/FS project will eventually sell out?

Re:Not gonna happen.

[Anonymous Coward]

> The world is moving toward software as a service. JBoss is positioned to be the king of that world.

How so? Marc's stunning business acumen? They have a piddling little consulting service, and it's not like their app server doesn't have credible competition from the likes of IBM and Bea.

I can already hear the whining

[blueZ3]

Wah wah "evil corporations" "poor workers" "outsourcing" blah blah blah.

One thing that seems to be overlooked is that with productivity rises, it takes fewer employees to do the same amount of work. The same is true after a merger, where it's redundant (no pun intended) to have two shipping departments or two sales forces.

I've been laid off several times in the last six years (once on Christmas Eve), and it's never been a big deal. I'm not saying it's been "fun" but if you have a rational savings plan to build a contingency fund, you should be able make it during the times you're laod off. I have sympathy for folks who are losing their jobs, having been there myself, but I also know this isn't the end of the world. I hope they do, too.

  You can look at a layoff as a crisis or as an opportunity. Your choice.

Re:Wow

[drinkypoo]

I think Oracle is doing something 'wrong'. I just don't think they're doing anything illegal - which has nothing to do with right and wrong, just what is or isn't convenient for society and the Powers That Be(tm).

Re:Not gonna happen.

[mbowen]

How do you get bigger than IBM and Oracle with JBoss? Simply because it's technically superior? You have to have superior sales and marketing to be a superior product, and only products make money - not technologies. JBoss needs to compete in the marketplace and it won't do so just because it was made by some cool guy named Marc. One day all of you open-source weenies are going to realize that the world doesn't run on GPL.

But.... JBoss is Open Source

[tux_fairy]

So I don't understand people's fear. If Oracle buys it and turns it into crap, you can always fork the latest source prior to the acquisition, right? Am I missing something?

Re:The JBoss deal is about Hibernate...

[fupeg]

You know this same thought crossed my mind too. However, on one hand Oracle has an app server and Oracle has a EJB3-compliant O/R technology. I'm not really sure how well their app server compares to JBoss, but Toplink certainly compares pretty well to Hibernate. Hibernate does have more mindshare than JBoss, but if you compared all the technologies in Oracle's middleware suite to JBoss equivalents, it seems like the one place where Oracle would stand up best is Toplink vs. Hibernate. Maybe what would be more valuable to them would be having the Hibernate guys, particularly Gavin King, as part of Oracle. That would give them a lot more influence on the future of EJB persistence and even JDBC.

Oracle will not practically own EJB3 persistence however. Don't forget about Kodo, a recent acquisition of BEA. They've had the best JDO implementation and now have an EJB3 implementation based on it.