Back to Basics for Sun Software (Continued)
Java in Context
Q: Let's talk about where you see Java as a language heading in the future.
Fowler: I don't think of Java as a language. Obviously, it has a language construct, but Sun's focus has always been to enable network computing. What's really behind Java and our interest in extending it is enabling the possibility of writing, deploying, and running applications across the broader network.
A long time ago, we saw—as most people did—that many devices and runtime environments will follow Moore's Law and get richer in semantics. To take advantage of those things, we would need some kind of runtime and development environment, but we weren't concerned with the wonders of garbage collection and the like. Those are pieces of [the picture], but the driving force goes back to Sun's roots in the workstation environment—putting the Internet and TCP/IP on a workstation.
Bill Joy's interesting notion was that most of the time when you use a computer, what you're interested in is somewhere else. This is important because it helps explain our future direction. We're trying to enable network computing for end-to-end architecture, from identity all the way to the back-end data center. So, it's back to "the network is the computer."
Q: To achieve that, you need a standard programming environment, right?
Fowler: Yes. Part of our vision always has been to provide something multiplatform, multiruntime, and open, so that people can use, develop, and deploy applications in lots of places. Before Java, we had standards in the Unix environment for developing applications, such as POSIX, but for the most part application development was fragmented. You wrote for specific APIs on specific platforms controlled by specific vendors, and that was the way it was. Consequently, a big part of the Java message is to create a level playing field for developers and create economic disruption by breaking the stranglehold of companies that produce specific APIs. This creates a larger possible market for people doing innovation, whether they are small companies or big companies—us and others.
Obviously, that makes our lives complicated. The traditional model is that if you invent something, you keep it entirely to yourself. We didn't do that because we've always believed in open standards and things that create a total available market—or a larger total available market of devices, people, and technology that connect to the network. Creating a large total available market is good for Sun Microsystems and its partners.
Q: What is your platform proposition?
Fowler: Being in the platform business is a long-term game. People see Java as an overnight success, but the reality comprised years of slogging away, getting things to work better, getting the performance better, making sure the security was there, evolving APIs, adding capabilities, porting to additional platforms, and so on. When you're in the platform game, you've got to be in it for the long term, and you've got to keep going, and you've got to make investments year after year. A lot of what you see today is a reaffirmation of [principles] that we talked about from the very genesis. I think that's good news because it takes long-term determination to get things done, to provide a stable environment for developers.
Q: Sun has been talking a lot recently about Linux, which seems like a tangent to the long-term strategy story of Java. What is the connection?
Fowler: We view Java and J2EE and the programming environments as the long-term story. The operating-system platform is an important detail, but we're trying to free you from all that. Having a couple different operating environments is useful, and right now, the press is focused on the operating system. We would love to get back to what we think is the core principle about Java and Sun One.
Q: But haven't you brought that on yourselves?
Fowler: We've amplified it, but a lot of comes from competitors and the press. Our internal strategy is not around "evangelizing" Linux or even Solaris APIs. Our strategy is "evangelizing" Java. That's the programming environment. That's where we want people to go.
We understand that we also make it possible for people to go somewhere else. It's part of our mantra that Java is open, and obviously there's a risk there. But that's where we think computing ought to be going, and Linux is a part of that. Linux is an exciting and powerful initiative against Microsoft. When you have a common developer runtime environment such as Java, you can span all these different things. When you add in the desktop and the server, Linux will grow market share against Microsoft, and that is an interesting development.
Back to top
|