[Originally Posted 19-May-2004]

In a half-assed attempt to regain customer loyalty BEA Systems (makers of the behemoth J2EE product WebLogic Server 8) have announced they’re taking the core of their IDE offering – WebLogic Workshop and making it available via Open Source. This was, of course, on the heels of BEA’s stocks dropping by almost 1/4 their value in one day after announcing abyssmal earnings.

Puffed-up shirt, BEA CTO Scott Dietzen, is saying that J2EE development is “too sophisticated” for developers. So, to make your development work as a J2EE developer easier they’re going to give you WebLogic Workship for FREE! This will “[allow] our customers to easily migrate a Tomcat app to a WebLogic container – a seamless upgrade path if and when they want it and there is a business case for it.”

So, this means that you can get their free IDE, and then get locked into their J2EE server as a result.

IBM has found success with this. They caught onto the Open Source trend and release their Eclipse IDE. Of course, the more you use it the more you find that their tools try to point you to using the commercial IBM solutions they offer (like Rational Rose). But IBM is IBM and of course people buy into whatever IBM offers. Just so you know, I think Eclipse sucks. That’s why I use JBuilder still.

Obviously no one at BEA can see the forest for the trees, so I’ll point out the obvious problems at BEA for them:

– Problem one: BEA
BEA has become too large and to fat a company to compete in the current economy. They have set up a business model worked fine in 1998, but in 2004 people are more shrewed with their software technology dollar. They’re not interested in spending US$10k per CPU to run a J2EE server that doesn’t include support of any kind for that price.

– Problem two: WebLogic
WebLogic is – quite simply – a pig. It’s a stinking bloated ass of a J2EE server. It’s failover and clustering facilities are difficult to configure. It requires *enormous* amounts of processor horsepower under load. Under the hood it’s still using it’s old proprietary T3 protocol in it’s processing core libraries because no one left at BEA understands it enough to tear it out at the roots and start over.

– Problem three: the WebLogic tech ‘support’ model.
I’m sorry but few people are brave enough to say this, because it’s not ‘politically correct’, but *I* will make a stand. I will defend BEA’s right to do business as they see fit, but when I pay for a support contract and the support under that contract gets farmed out to a country where english is spoken as a THIRD language, I’m out. I will take my money elsewhere. I just tore BEA to shreds over this in a survey about their support. Scripted responses, slow turnaround times, and a lack of comprehension of my questions (due to an inability to speak english over a fifth-grade level) are NOT an appropriate expenditure of tech support dollars, in my opinion.

How BEA can fix these problems.

WebLogic Server 9 should be open-sourced (or at least heavily discounted):
WebLogic 9 offers *nothing* that JBoss or Borland Enterprise Server doesn’t offer. In fact I’d go so far as to say it doesn’t offer as much as the other two offer. However JBoss is free (you pay for support). BES is half the money and twice the performance, functionality and ease of use. You should look at the BES administration tools – they’re a model of simplicity. Going forward, BEA should provide some core component of WebLogic Server 9 for free – zilch – zero, for commercial use. And offer the current ‘commercial’ version for a price in the US$2000 range per *SERVER* (not CPU – that’s just a sick joke). Perhaps offer a “Foundation” version that doesn’t provide clustering and HA support, for example.

– WebLogic should pull support *back into the US for US customers*.
Go ahead – charge 90 bucks per support incident. Just make sure I talk to someone who can speak english as a *first* language and truly knows WebLogic inside and out when you do.

– Reduce headcount.
Yes, that’s a nasty thing to say, but under their current business structure I see them continuing to post losses unless they make some serious headcount changes *or* get the above support options in place and put the support contract work under the direction of their Professional Services groups that they have in satellite offices around the country.

BEA is facing some tough times – if they want to remain viable they need to provide *value*. And right now, they don’t.