RealTime IT News

David Rivas, Mobile Systems Group CTO, Sun

David Rivas Sun Microsystems' Java 2, Micro Edition platform is among the most popular programming platforms in the industry.

A scaled down version of Java 2, Standard Edition (J2SE), J2ME is used to create applications on PDAs, mobile phones and embedded devices.

There are 350 million Java-enabled handsets today, a number that's expected to grow to 1.5 billion in the next three years, according to research outfit Ovum. And with J2ME market penetration of 75 percent for 3G phones, the software maker is dominating the environment away from the desktop.

But despite these figures, Sun is having some trouble getting the types of developers that they are seeking. Also, the applications most frequently developed on the J2ME platform are entertainment programs, not the enterprise tools J2ME is suited to accommodate.

End-to-end applications, those that tie the database and other back-end functions to the application, are not as prevalent as Sun would like.

David Rivas, Sun's consumer and mobile systems group CTO, wants to increase Java's presence around the world. He recently discussed some of the issues facing J2ME and what's next with internetnews.com.

Q: J2ME is one of several development platforms for mobile devices. How does Java stack up and what sets it apart?

One of the benefits of Java, and one of the reasons we're seeing more and more operators determine Java is the direction they want to lead their developers, is that it goes beyond what operating system you are using. What really matters to a developer, what really matters in getting applications out there, is the availability of an end-to-end programming platform. So this question of, "Is it Symbian, is it Palm, is it Linux?" or take your pick of operating systems, becomes less of an issue. Take that decision out of the realm of what a developer needs to worry about, what we're seeing is that more and more applications need to be downloaded to a broad collection of devices. It's not actually scalable if a binary application is particular to a particular operating system, because you'll never be able to keep up. You have a common platform out there in Java that actually solves that problem quite nicely.

Q: The learning curve for J2ME is pretty steep, what's Sun doing to address it?

We've been increasing the capability of our toolsets, it's winning awards all over the place. It is one of the top developer platforms for wireless developers; it supports a collection of features necessary to do really good development. There's always room to improve with tools, I believe that for any tool that's out there. Developers will always want more and one of our responsibilities is to give them more, as well as partner with some of the tools vendors out there.

I think one of the things we're beginning to see is developers take to the notion that they could actually develop a markup language -- to make it quick to develop -- where they provide the engine or buy the engine from a third-party that runs in Java that renders that markup language. It's kind of the best of both worlds and that's something unique to the platform we provide. One of the traditional problems with markup languages is that once it gets deployed into the device, that's the only markup you could use, and whatever shortcomings there are with the markup language, the developer is stuck with that.

In the Java model, where the rendering tools for the markup language are actually the Java application, one's in a position to hire or write or buy a markup rendering engine and then you have an opportunity for developers to write in that markup; that increases the productivity of the developers, you decrease the skill level required for application development. It's a much more natural environment.

Q: It seems that e-mail is still the killer app for mobile devices, is there anything on the horizon that will supplant it?

I guess I don't really know for sure what's next, which is perhaps indicative of us not quite being there yet. I think the last couple years it's been a lot about the destruction of barriers to entry; we're now at a place where it's actually possible for a lot of applications that are currently available at your desktop, or currently available on the Web, to be provided in a mobile environment. Which one of those, or collection of those, is going to drive the next killer app, we don't yet know.

Q: What's next for J2ME?

The big issue for us is to continue to drive the number of deployments. The goal is to go from 250 million Java handhelds to 1 billion Java handhelds and to do that we not only need to increase the amount of developer activity on the platform, get the message out that now is the time to make your applications, but we need developers to increase the end-to-end development. We've had lots of success with those applications but they make limited use of the network. The challenge and the opportunity is for the developer, in conjunction with the operators in the enterprise, to bring the value of that network to the device itself. One of the ways we're facilitating that is we're making it easy for developers to use the complete Web services infrastructure that is necessary to actually bring the value of the network out to the device.