At BEA Systems Inc.'s annual eWorld conference last week, CEO Alfred Chuang spoke with Computerworld about the application server market space, his company's competition, recent trends toward Linux and the open-source software movement.
Everyone talks about application servers becoming a commodity. What are your thoughts?
I wish it was a commodity. I think there's a perception in the marketplace that commodity is bad. So far in the software arena, everything that has become a commodity has driven tremendous adoption. Microsoft Windows is a commodity, but nobody relates commodity to not making money in their case.
I think it's just a philosophy thing. Do you want to have mass proliferation of a technology? If you want to do that, then whoever is providing that technology has to make it a commodity. I think BEA is doing some significant [work] to make part of our technology a commodity. If you look at our [WebLogic] Express product, which was announced recently, that's $495 per CPU. It's cheaper than most desktop software. It's cheaper than Microsoft Office Standard. With our free Dev2dev subscription, you can literally develop your own very powerful, dynamic Web site, commerce Web site, for $495.
How can your company thrive when others such as Microsoft Corp. and Sun Microsystems Inc. are building or including their application server technology with their operating systems?
I think having the market recognize that the application server, especially the J2EE [Java 2 Enterprise Edition] application server, has become the standard in the marketplace is a great thing. People have discovered that nothing coming from Microsoft has been free. I am a Microsoft customer, and I don't expect that it will be free. It's just that the point of being charged is going to change. To combat that, you have to have a very smart pricing strategy to really know how you can provide the most value to the customer....
So far, if you look at our financial performance, our adoption in the marketplace, the number of users on Dev2dev, the number of licensed seats that we have sold, the number of customers that we have, how many we're adding, we're heading to the right direction. But there's still a disparity between the potential of the enterprise computing field and where we are.... BEA will have to do more. I think we have to be even smarter in pricing.... We have shown the world that we can triple our profit in this year. Clearly, we have no problem generating a margin by selling a technology that people thought had been commoditized.
At your conference, there were a lot of potshots tossed at IBM. What worries you most now about IBM?
We are in a very depressed economy, and we have been in this rut coming on three years next month. Nobody thought that it was going to last this long. In a contracting economy, you have players that all have to fight for a non-growing amount of revenue. So obviously, that puts pressure on everybody. We may take some potshots at IBM, and I think IBM takes significantly more potshots at us. There's a divide of two camps. One camp is the camp that says buy everything from us -- the mouse, the keyboard, the LCD monitors, the laptop, the midtier server, the mainframe, the storage, the software, everything. And then you have the other camp, and you say, 'Well, you buy this best combination and that's going to work very well together to suit the customer better.'
I think there's a conflicting set of interests between the closed, monolithic camp and the open camp.... I just believe that I'd much rather be on the 80 percent side where you have much more to work with.... Most of the computing money is spent in the enterprise, in the software, so I think we are in the sweet spot.
Do you consider .Net to be as big a competitor to you as IBM's technology?
Everything that Microsoft does I take very seriously, because I know that they are our ultimate competitor into the future. Microsoft says enterprise computing is where they want to be next. In short-term reality, there's almost no competition that goes on [between us], because our channel model is very different, and there's really no .Net server. You can't compete with a hypothetical thing when the customer really needs something to deploy. We sell real stuff.
I think ultimately the channel model makes a huge difference. Our customer hasn't changed very much, but they still need a lot of hand-holding. They want tender loving care. They want somebody to call.... If you look at our maintenance revenue growth, you see that our customers are getting even more loyal to us because they are so concerned with the stuff that they have in production, it becomes so mission critical to them, they've got it running all the time.
What Linux trends are you seeing with BEA software?
Huge adoption curve climbing very fast for BEA over the last six to nine months. A lot of focus in the financial services marketplace, where there's a lot of experimentation and initial deployment going on with Linux on Intel. And I think the motivation in that arena is simplification and cost reduction, so they are looking to buy significantly less expensive hardware.
What's the breakdown of platforms on which BEA software is running?
About 50 percent is on Sun, and about 23 percent, 24 percent is on Hewlett-Packard. Hewlett-Packard has both Intel and non-Intel platforms in there. And then it drops off pretty quick. IBM hardware, I think, is 5 percent or 7 percent. In some countries, we sell a lot of IBM's hardware.
What about the Linux operating system?
Linux is around the 15 percent to 20 percent range, which has climbed pretty quickly.
From what in the past year?
Do you think Java should be made open source?
I think there's a miscommunication in the marketplace. Open source doesn't cure hunger. Open source clearly is a great, terrific way to help proliferate knowledge, and I think open source becomes very, very useful especially when you have technology that's supposed to be open and commoditized and simplified, and it has gone in the wrong direction. Unix is a very clear example. Unix started even simpler than Linux in the beginning. It was a tiny little kernel, and you had applications around it.
Oh my God, after 32 years, we have bundled the kitchen sink into it. The binary is so big, and there's so much stuff that you're never going to need. So it has gone in the wrong direction. I think Linux came at the right time. Open source in that perspective is very, very useful. Now give it another 20 years or 10 years where at some point, some part of the application server should be open source.
What piece do you think would make a good candidate for open source?
I think database drivers. Things that application programmers don't program to anymore would be the right candidate.
Do you believe in making any of your own code open source?
Right now, we certainly are not leaning toward that direction because our customers demand our code to have utmost integrity. Open source means that not only can people see your source code, they can contribute to the source code itself. So in that fundamental process, you have to lose control. You can't have 100 percent control and 100 percent of open source at the same time. You cannot have the best of both worlds. Something has to give in that process. Right now, our customers want integrity. Our customers want control.