InfoWorld editor at large Paul Krill met with Mike Milinkovich, executive director of the Eclipse Foundation for open source developer tools, at the EclipseCon 2007 conference in Santa Clara, Calif. to get perspectives on Eclipse and issues such as the recent open-sourcing of Java.
IW: We talked this morning for a few minutes about the theoretical Eclipse 4.0 platform. What do you think might be in that? You're already covering about 80 different technology areas, so what's left to add to it and what might you expand on in Eclipse 4.0? (The current release of Eclipse Platform is version 3.2.)
Milinkovich: The idea of doing an Eclipse 4.0 is mostly focused on the platform, so it's not so much about expanding the coverage of Eclipse, as you put it. It's about the core technology that those 80 different projects build on top of. And frankly, at this point, I don't know, we're really just starting the conversation about what the community thinks would be useful additions.
IW: What constitutes the core platform?
Milinkovich: Currently, the Eclipse project consists of a number of different frameworks, but they can be basically grouped into four different areas. There's Equinox, which is the basic core runtime for Eclipse. There's the tools platform, which is the basic infrastructure for building Eclipse-based tools. There's the JDT, or the Java Development Tools project, which is the Java IDE that most people think of when they think of Eclipse. And then finally, there's the PDE, the Plug-in Development Environment, which is the tool that you use for creating Eclipse plug-ins.
IW: When might there be an Eclipse 4.0? I think you said not for a couple years.
Milinkovich: Yes, not for a couple years. This is the very beginning of the conversation. The first question is, does the Eclipse community even think that doing a 4.0 would be a good idea? So these are very, very early days, and it's very exploratory at this point.
IW: Do you think it is a good idea?
Milinkovich: I think probably, yes. On balance, it's a good idea.
IW: Why?
Milinkovich: Because you don't want to be stagnant. You want to make sure that you're evolving your platform. In any software project, it's important to make sure that you're constantly working to keep the technology fresh and current. But we need to get feedback from the community before we make any decisions.
IW: Eclipse has been around since 2001 and was broken off into its own organization in 2004. What would you say is the value of the economy around Eclipse? How many products are being sold based on Eclipse?
Milinkovich: Let me preface this by saying that these are my somewhat informed guesses. We don't have hard numbers on a lot of these different things. But in terms of the number of products that are built on Eclipse, it's clearly in the hundreds if not, perhaps, even in the low thousands. In the embedded space alone there are probably hundreds of products. I have a Google Alert for Eclipse embedded, and I'm constantly getting bombarded by new products in that space. In terms of the overall dollar value, I don't really know, but I'm pretty confident that it's well over $1 billion, in terms of the size of the entire ecosystem. And I'd love to find somebody who's willing to do a free study to give us better numbers.
IW: You mentioned this morning that Eclipse may become as big with runtimes as it has been with tools. Could you talk about that?
Milinkovich: This is really around taking the Eclipse platform, which is this OSGi-based (Open Services Gateway initiative) plug-in model, and starting to see people build applications with it rather than simply tools. If you go back to Eclipse 3.0, that's the first release, this was in 2004. That was the first release was shipped, [featuring] the Rich Client Platform. The first Eclipse runtime at that point was on the desktop. Since then, we've had the embedded Rich Client Platform Project to bring that technology to devices, and we have several different projects at Eclipse that are bringing the same technology to the server. So we see a lot of growth and a lot of interest in leveraging this technology in applications and in products.
IW: The two big holdouts from joining Eclipse are Sun Microsoft and Microsoft. Do you see those two companies eventually joining Eclipse?
Milinkovich: Well, first of all, the door is always open. We'd love to have them. There haven't been any recent discussions with either group, so I don't think there's any change with either company. I think the one thing that's happened recently is Eclipse joined the Java Community Process. But there hasn't been any interest in Sun participating in Eclipse that I'm aware of.
IW: Could you just reiterate the importance of OSGi to Eclipse?
Milinkovich: OSGi is the open standard that we base the Eclipse runtime on top of. So Eclipse Equinox, for example, is an implementation of the OSGi specification, and it is the heart of everything that we do at Eclipse. One of the things about Eclipse [that] makes it a little bit unique is that we have an architectural constant that goes through all the projects at Eclipse. Everything we do at Eclipse is about building plug-ins, which are effectively OSGi bundles, and so there's this very strong link between everything we do at Eclipse and fostering greater adoption of the OSGi specifications. Basically everything we do at Eclipse is extending and growing the OSGi community.
IW: What is the purpose of the OSGi developer conference that's happening concurrent with this one?
Milinkovich: There are so many interrelationships between the two groups, and there's so much interest within the Eclipse community about OSGi. We decided that there was going to be enough overlap in content that it would make sense to have them come here, and they were definitely interested in doing that.
IW: I think I read that there about 1,300 attendees at EclipseCon.
Milinkovich: Yes.
IW: I've been looking at attendee badges. It seems like most of them are from different commercial software vendors. Is that the makeup of most of the attendees here?
Milinkovich: The demographic, if you will, for EclipseCon is developers either working on the Eclipse platform or using the Eclipse platform to build products. And so that is probably the bulk of the badges you'll see here at EclipseCon.
IW: What are the goals for your organization in the next year or so? Do you expect to just complete some of the projects or continue with a lot of the projects you're going for?
Milinkovich: We are an open source community, so it's about the code. And so the goals for our organization this year are primarily around taking the projects that are just getting started and making sure that they get their first releases out the door, and [with] the projects that are more established, making sure that they continue to grow their group of adopters and make sure that they continue to build product-ready technology The big event every year at Eclipse is when we ship our annual release train, and this year the Eclipse Europe release train going out in June has currently 22 projects.
IW: And that's about providing unified releases of everything at the same time?
Milinkovich: It's not a single large software release. It's 22 Eclipse projects shipping on the same day. And that's still very valuable to the Eclipse community because they have been at least somewhat tested in terms of their ability to load up into the same workbench and work together, but the primary reason why we do it is so that commercial adopters of the technology can have all of the projects that they need to build their products.
IW: Do you think the concept of open source has reduced or raised revenues for software companies? I've had it mentioned to me that open source reduces revenues for everybody.
Milinkovich: On the whole, I would say that open source has been a very large benefit for software companies and the software industry as a whole. Yes, for any one particular company you can say their revenue went down. But for every company like that, you can find other companies whose revenue has gone up. People have made some very large businesses around open source. So I think that it's been a good thing for the industry. It allows people to build more innovative products because they can focus on their differentiating value, the thing that matters to their customers, and spend less time building the infrastructure underneath them. I think it's actually accelerated the rate of innovation within the software industry. And the other thing I'd point out is that there are a number of very large companies whose business model would basically not work without open source.
IW: What companies would those be?
Milinkovich: Google would be a good example. The software costs alone of outfitting however many thousands of servers Google has would be astronomical if they didn't have the ability to create an open source stack to run on those machines. So I think that's probably the best example, or the most obvious example.
IW: What's Eclipse's take on the open sourcing of Java?
Milinkovich: I think it was a good move. It's good for Java to become part of the mainstream of open source development. One of the knocks against Java for the last couple of years has been because it was a proprietary technology, that it was not something that many free software and open source people were interested in using. By making it open source, it brings it into a broader community and reinvigorates the ecosystem around it, and it hopefully will attract even more developers and greater innovation.