An Exclusive Interview With Mike Milinkovich Of Eclipse Foundation
The importance of adding companies such as BEA, Borland, and Computer Associates to our board cannot be overstated
By: Jeremy Geelan
Jul. 18, 2005 10:00 AM
Mike, thanks for agreeing to talk to JDJ and bringing us all up to date with Eclipse since our last interview with you. The best way is probably just to fire off questions and allow you to answer without getting in your way!
JDJ: Overall, how's the independence from IBM going? Since most of Eclipse's committers were IBM employees when you first went independent, how is building a community around Eclipse that is not so IBM focused going?
MM: It's going really well. Frankly, it's going much better and much faster than I had originally anticipated when I started the job. In many ways we have accomplished in the past year what I had expected would take two years or more.
First, the importance of adding companies such as BEA, Borland, and Computer Associates to our board cannot be overstated. Each of these companies competes fiercely with IBM in the marketplace. Each is making million dollar plus investments in Eclipse ($250,000 per year in dues, plus a minimum of eight developers). Each did their own analysis as to whether the Eclipse Foundation was truly independent. And each joined.
Second, the number of projects led by non-IBMers has increased dramatically over the past year. In terms of top-level projects, we have gone from three projects in which two were led by IBM to a total of eight projects with two still being led by IBM.
Third, the number of committers working on Eclipse projects who are IBM employees has steadily dropped over the past year from roughly 75 percent to just over 50 percent. The number will soon drop below 50 percent. This decrease has been mostly the result of increasing the total number of committers. We certainly do not turn away good people from IBM!
In fact, I would also like to recognize the investment that IBM has made in Eclipse. They started this adventure and their continuing investment remains impressive. I also think that the Java community as a whole should recognize the wisdom and strategic thinking shown by IBM in working to establish the Eclipse Foundation as a separate entity. IBM has truly demonstrated how to create a community.
Eclipse is truly really completely fully and utterly independent. Anyone who says otherwise has an agenda.
JDJ: How has the assimilation of so many new strategic developers after EclipseCon gone? Is the "plumbing" in place to handle this influx? Are you making adjustments to better handle large groups of people coming on board?
MM: I don't want to sugar coat things. We are having our growing pains as we start up all of these new projects. We need to help get these projects off to a good start and our processes and people are stretched in doing so. However, it's not just the new strategic developers who are causing growth. We have been receiving project proposals from many different directions.
The reaction from the Eclipse community has been outstanding. People with experience within the Eclipse community are stepping up to help us refine our processes and work with the new project leaders and committers to help them get started.
Back in May we had our latest round of Eclipse Council meetings and it was by far the best set of meetings we've had. We were actually debating hard topics like what does it mean for a project to achieve "Eclipse quality," how can we work toward being an even more stable and predictable open source community to further encourage commercial adoption, and what are the tangible things that existing projects can do to help new projects get started. I think it's a very good sign when our community leaders are constructively discussing the tough issues.
JDJ: How much of the themes and priorities specified on your Web site -www.eclipse.org/org/councils/themes.html - were you able to accomplish in the 3.1 release?
MM: All of them. None of them.
The themes are not intended to be deliverables. They are intended to provide high-level focus to many different projects at Eclipse to concentrate on similar topic areas and help our entire community move forward. I predict that we will be working on some of those themes for years to come. Certainly at Eclipse we will never step working on "Being a Better Platform." We take the construction, stability, and evolution of the Eclipse APIs very seriously. We want developers to feel comfortable relying on our interfaces.
There are a couple of things about the Eclipse community that are very unique in the open source world that bear repeating.
First, the community is committed to quality and stability. The committers on our projects work very hard to maintain API stability from release to release. Anyone who has ever done this knows that achieving such a goal is a non-trivial effort. But the community honestly believes that API stability is one of the key elements that has made Eclipse so successful.
The second is that we are very focused on predictability. We want our technologies to be adopted commercially. If you want companies to bet their own product plans on an open source project, you need to deliver on schedule. If you look at the Eclipse Platform project, the 3.1 release will be the fourth release in a row where they have hit their dates. If anyone wonders if you can deliver open source projects with several million lines of code on schedule using developers from multiple companies at multiple sites, Eclipse is the project that erases any doubts that it can be done. It is a real testament to the amazing people involved in the project. Pound for pound, I think that Eclipse has some of the best software engineers on the planet.
JDJ: Will the WTP be included with Eclipse by default when it's done?
MM: I don't know. We are still discussing cross-project packaging options within the community.
JDJ: Tell JDJ's readers about the RCP initiative. How has the uptake of the RCP been going? Are you seeing a lot of interest?
MM: If you are a developer interested in desktop Java development and you have not yet looked at the Eclipse Rich Client Platform, you really need to. Eclipse RCP offers a great platform for developing managed applications with a rich user experience. It's doing more to generate interest in Java on the desktop than any other initiative.
We have seen enormous uptake of RCP. I wish we could talk about all of the projects we've seen. One of the most intriguing is the NASA Space Mission Control used for mission planning for the Mars Rover. The most ambitious is the IBM Workplace Client initiative. Macromedia's recent announcement states that they will be building their next-generation Rich Internet Application (RIA) environment on Eclipse RCP.
Reader Feedback: Page 1 of 1
SOA World Latest Stories
Subscribe to the World's Most Powerful Newsletters
Subscribe to Our Rss Feeds & Get Your SYS-CON News Live!
SYS-CON Featured Whitepapers
Most Read This Week