With EclipseCon taking place this week in Virginia, there's an awful lot of neat projects to explore and talk about in the popular tools integration platform. Perhaps the easiest way to take a stroll through the Eclipse ecosystem's latest developments is to look at the winners of the Eclipse Community Awards.

Winner of the Most Open Project award was the Eclipse Communication Framework. Mustafa K. Isik, a contributor to the ECF project, attributed the win to the work of Scott Lewis, the projects lead. "IMHO, this is first and foremost [Scott's] achievement. [Scott has] always set the tone and course for openness and acted accordingly, serving as an example for us all," wrote Isik in a posting to the ECF mailing list.

The Most Innovative New Feature award went to Eclipse Code Recommenders. That project is currently at version 0.5, but is pushing quickly towards creating a revolutionary new method of code completion. Code Recommenders is all about teaching developers how to use essential APIs directly in the IDE, in the code. The project uses extended javadocs, smart bug detectors, stacktrace search engines and intelligent code completion to give developers highly relevant suggestions.

Outside of Eclipse, the Community awards also recognised projects and products based on Eclipse. This year's winners clearly show that Eclipse users appreciate better debugging and rapid prototyping tools.

Best Developer Tool honrs went to the Chronon Time Travelling Debugger. This tool that comes with the Chronon test system for Java applications allows for some very interesting debugging possibilities. Chronon calls itself "DVR for Java," meaning applications can be run and recorded, then played back to find root causes for bugs. The debugger actually lives inside of Eclipse, and it was this integration that won the award.

Best modeling product was MaintainJ, which is praised for its ability to turn complex Java applications into more comprehensible models. And the award for best application went to the Justinmind Prototyper, which allows developers to quickly design Web application experiments.

I've had the privlidge to attend EclipseCon out here in Sunnyvale every year for the past six years. That changes next year: EclipseCon North America is moving to Virginia. Here's the announcement from Ian Skerrett:

EclipseCon North America is moving! The Eclipse Foundation is pleased to announce that EclipseCon 2012 will be March 26-29 at the Hyatt Regency in Reston, Virginia, outside Washington, D.C. We are excited about the great new location for this conference, and about moving to the east coast of the U.S.

Doug Schaefer of Wind River is Program Chair, and we expect to announce the rest of the committee shortly. The new website is coming soon, as well as the Call for Papers. Follow us on Twitter and Facebook for updates.

Be sure to save the dates, and we look forward to seeing you in Reston next March.

Ian Skerrett
Eclipse Foundation
EclipseCon 2012 Conference Chair

On the heels of the announcement that Oracle will begin the process of moving the Hudson CI project to Eclipse, Koshuke Kawaguchi, the creator of Hudson, posted on his blog that he was 'surprised' by the announcement. Instead of saying it's a good idea or bad idea, he emphasized he tried speaking to Oracle about such a move in the past but was rebuffed. Then, on the thread that emerged, Ian Skerritt of the Eclipse Foundation admitted a "communication oversight" but said the move to Eclipse would be a good thing. Oracle's Ted Farrell also weighed in, saying the company proposed the move to Eclipse in December but was rebuffed by the Jenkins. There's a lot of bruised feelings and jockeying for moral high ground going on, so I posed this question on Kawaguchi's blog: "So let me ask point-blank.. will the Hudson and Jenkins sides be able to put aside their differences over project leadership and work together, bringing the forks together again? Are talks happening beyond the exchange about poor communication on this blog? That’s what people want to know!" Of course, there was no direct response; instead, there was much back-and-forth about licensing issues, and the rights under the various licenses involved. It's clear the Jenkins crew believes it is the rightful heir to the project, as its creator is on their side. Yet Oracle owns the Hudson name from its acquisition of Sun. The initial squabble was a difference over rapid releases (Jenkins side) vs. stability (Hudson side). By bringing Hudson into Eclipse, which has a great record of building community around open-source projects (rapid releases) that are then productized (stability), both sides should be satisfied. Oracle's announcement is but a first step, but one that should properly be explored, and ultimately chosen. Let the two sides check their egos at the door and work for the betterment of the community by bringing the forks together and creating something that both keeps pace with changes in the field and provides the stability enterprises require to deploy such software.

I had the delightful opportunity to speak with Mike Milinkovich, director of the Eclipse Foundation, earlier this week. He's in town for EclipseCon, which is where we chatted. Eclipse is rolling along nicely, and he said that this summer's Indigo release is going to be the biggest gift Java developers have received from the Foundation in about 5 years. The inclusion of WindowBuilder, recently made free by Google's acquisition of Instantiations, should make Eclipse a complete end-to-end IDE for the first time. While many other GUI building tools have been brought to Eclipse already, WindowBuilder is a remarkably mature offering with full round-tripping capabilities. Milinkovich also took credit for bothering Google until they opened the software to the public. As we speak, Google and others are working on preparing WindowBuilder for the Indigo release train.

Here's a transcript of some of our interview:

You seem to like WindowBuilder a lot

WindowBuilder has been around for a really long time. I remember using it when I was a Smalltalk programmer. That technology has been around for a long time. They had several hundred-thousand paying customers when they were Instantiations. This is not something that's starting out, it's stable, it's mature, it's well known amongst Java developers.

As an Eclipse project, however, it's just starting. [March 21 was] the first day they had a build you could download from Eclipse. I'm very confident that when it shows up in the indigo packages, people will love it. The team is working extremely hard to make that happen. Getting it into Indigo will set a new land-speed record for a project going from proposal to acceptance.

As always, this is the most technical show I attend all year.

The keynotes are great. Did you see the Watson keynote? This is a very technical crowd, and it was a fairly deep dive into IBM's Watson. All the feedback I saw on Twitter was positive and there were lots of really good questions at the end. [On March 22] Mark Reinhold is talking about OpenJDK. I think that's going to be of great interest to everybody in the room. Then there's a panel afterwards. I'm sure we'll get some good questions. [They did!]

What's up with the OpenJDK? You're on the governance committee... what do you want out of this project?

We could be more transparent, faster. We made a lot of progress. We've gotten a lot of good feedback from the early drafts, and I expect there will be more drafts. The OpenJDK governance body is focused on redoing the by-laws. It's moving. People from the outside don't see any progress. But the message is "progress is happening, stay tuned." There will be more content shortly.

You're also representing Eclipse in the JCP. How would you assess Oracle's overall approach to Java?

As I understand it, Java 7 is scheduled for release late this summer, and as far as I know they're on track.Under Sun, the code wasn't moving forward. One of the things about Oracle's stewardship of Java is that it takes time to get going, but they are actively investing in moving Java forward. Sun was using the deadlock at the JCP as a convenient excuse to save money on a lot of engineering resources. The pace of innovation in the Java platform is going to get a lot better. In terms of Java as a language, some innovation is going to happen there, and around modularity. But there are a lot of things Java needs to do to innovate and to be more relevant to today's world. It needs to be much better and more relevant to Web developers

So, modularity is scheduled for Java 8. Will that modularity look like OSGi?

There is not yet an answer to that question. I made it really clear in Eclipse's vote on the Java 8 JSR, that if there wasn't room for OSGi to play within the modularity story for Java 8, that we would be voting against it. There is work going on around bridging that gap with conversations amongst the technology guys. In my view, it's a no-brainer. There is a lot of stuff being built on top of OSGi right now, and it's in nobody's interest to break that in Java 8.

The Eclipse Foundation and Google have gotten together to form Eclipse Labs, a site where developers can host their Eclipse-based projects. Eclipse Labs is based on Google Code, so it should be familiar to at least some open source developers. The Labs are to be more of a repository than anything else. Anyone can host Eclipse-based work there, but doing so does not make your work an official Eclipse project.

The Labs opened yesterday, but there are already some projects hosted there, such as redview, anyedit, and an integration plug-in for using Maven with Android.

It's almost over, so you'd better get yourselves to the Survey Page right now. The Eclipse users survey is always a wealthy source of information, so please be honest and give them your time. 

Ian Skerrett has the low down on how the whole logo process went down. Go vote right now!

PHP n' CDT... Also, Germans. That's what I'm taking away from the recently released results of the Eclipse user survey. They're available online for all to see. I certainly never knew that Germany was the country in which the most Eclipse users reside. Or, at least, of those users that responded to the survey. Naturally, ya'll responded, right?

1,481 people took the survey; Not exactly a large enough pool to yield certainty, but a healthy helping, no doubt. Traipsing through the Survey Monkey results, we find some new trends and see the continuation of others. One of those long-running trends is the continuing proliferation of Web frameworks: Over 60% of the respondants said they used "other" Web frameworks. GWT, Flash and Dojo were the most popular, however, running neck and neck in the teens.

New trends showed around the PHP Developer Tools, PDT. It came in as a firm second as the most used language tool, next to JDT, of course. The C tool suite came in a close third, however. Also, Ubuntu was the second most popular operating system for both development and deployment. That was not what I was expecting at all, despite my admitted Ubuntu fan-boyishness. I fully expected more Mac users to represent, and I never expected all the other Linii (is that the plural of Linux?) to be so thouroughly outnumbered. Put Fedora, Debian and other Linux distributions together, and Ubuntu still came up as the leader on the desktop. Red Hat, however, is a strong contender for Ubuntu's throne on the deployment side. Of course, Microsoft still reigns supreme in both client and server-side OS wars.

Of course, what's an open-source survey without open-source comments? Here are my three favorite slips from their suggestion box, all (sic) of course:

• make eclipse go to a big diet, let users uninstall also plugins from the base install (for example CVS I don't use anymore) and remove obsolete features (i.e. XDoclet support) or bloated ones like the WS Explorer, fix the update manager once for all, so many iterations and it still sucks, both in UI and functionality, a really bad advertisement for OSGi
  


• In my opinion Eclipse is great for Java community but as a IDE it is to complex for doing simple tasks on task oriented development process. Many plug-ins sounding great by there name and/or description but the installation could fail and kill complete update system until reset workspace and delete plug-in by hand. Other plug-ins wont respect shared configurations and mostly they wont work together with other IDE in the same project. Our team has the rule that every team member can use there own preferred IDE but if someone use Eclipse all other could run into problems and this slows down productivity of complete team.
  
  
• Eclipse is seen by many as THE Java IDE. I sincerely hope that the Eclpse project can expand this reputation to other (object oriented/functional) programming languages such as Haskell, Scala, Eiffel, ...

Pause a moment to look at this dialog box from the Eclipse STEM project. STEM stands for the Spatiotemporal Epidemiological Modeler. Just looking over this dialog box, you should be able to understand what I'm after here. Just how bad can this swine flu get? I certainly don't know the answer, but I do know that we can all take time today to model the worst and best case scenarios in Eclipse. This is a really fun toy, I have to say. It's somewhat complex, but the folks behind it have written some great guides to make sure you know what you're doing. 

First, you'll want to install STEM on top of Eclipse. You can download it here. Once you're up and running, the above dialog, which is relatively self explanatory, can be found in the terrific tutorial. There are a lot of other uses here, too. Have fun terrifying yourselves!

A friend pointed out to me yesterday that Ubuntu just isn't cool for developer workstations because all of the current and past versions halt at Eclipse 3.2. For those who don't know, Ubuntu is based on Debian, thus its users live out of repositories of tested and supported software. Unfortunately, this means that Ubuntu tends to be a few steps back on the releases of major projects.

Thus, Ubuntu is still distributing Eclipse 3.2. That's not good for people who want to use Ubuntu as a Java development desktop. It's also bad for those who want to program C or C++ with the now excellent tools in Eclipse. My friend pointed out that most Linux developers need C and C++, and that Eclipse's support for those languages only becomes interesting in 3.4. 

I e-mailed Mark Shuttleworth about this, and here is his reply on the subject:

Hi Alex, I have further information.

At this stage we don't have plans to bump Eclipse beyond 3.2 in the current development cycle. There is a community member who has made a start at some packages in a Personal Package Archive (PPA) at https://edge.launchpad.net/~pktoss/+archive/ppa but those packages are not ready for inclusion in Ubuntu and definitely won't make 9.04.

It's a few weeks' work to do Eclipse 3.4 well, and it hasn't been raised as an urgent priority within the community of folks who participate in Java work on Ubuntu, so [it] wasn't a focus till you mentioned it. If there's substantial demand for this, it could certainly be addressed for 9.10 (though I see 3.5 is due thereabouts, and it may be better to aim for that).