Eclipse Community Survey Results for 2013

The results of the Eclipse Community Survey for 2013 are now available. This is the sixth edition of the survey so it presents a nice snapshot of the changes that are happening in the Eclipse community and the wider developer community. A complete report that summarizes the results are available and as always we published the detailed data [ods] [xlsx] so you can do your own analysis.

Each year, there are a number of interesting trends that seem to emerge from the results. Here are some highlights from the 2013 survey:

1. Git continues to take-over. The adoption of Git as a developers primary source code management system continues to move forward and at the expense of Subversion. Interestingly, Subversion continues to be the #1 SCM but just barely. If you combine Git and GitHub usage, Subversion leads by only 1.5% points. Next year I am sure Git and GitHub will be #1.

2. Decline of Ant usage. For Build and Release Management tools, there was a significant drop in the usage of Ant. Only 38.8% on the respondents claim to use Ant compared to 50.8% in 2012.

3. Regional differences for cloud computing. This year we had a large number of respondents from Germany. This allowed us to identify some regional differences in the responses. One of the most significant diffrences were plans for cloud computing. Overall, 47% of  respondents had no plans for deploying applications to the cloud, compared to 59.5% in Germany. For those that were deploying applicaitons to the cloud, Germans top preference was to deploy to a private cloud (35%) compared to the overall respondents preference of deploying to Amazon (30.8%).

4. Perceptions of Web-based IDE. For the first time this year we asked about people’s experiences with web-based IDEs. Projects like Orion are investing and innovating in this new style of IDE so I thought it would be interesting to understand current perceptions. A large majority (72%) stated they had no interest in a web-based IDE or needed more information to form an opinion. It is certainly early days for web-based IDEs but I think these results show that there is a lot of education and evangelism needed to change developer’s perception of web-based IDEs.

5. Most respected companies in open source. This year we asked respondents to list which companies they respected for their open source contribution and participation.  This was a write-in question so respondents could write-in multiple company names and were not picking from a list.

The top 3 most respected companies for their open source contributions and participation are Google, RedHat/JBoss and IBM. Congratulations to all 3 companies, a well deserved recognition for the support they provide to the open source community.

6. Adoption of new Eclipse releases. For the first time this year there was a noticeable decrease in the adoption of the latest Eclipse release, Eclipse 4.2. In past survey results, 75%+ of the respondents would report they were using the most recent release of Eclipse, for example in 2012 76.9% were using Eclipse 3.7 (Indigo). This year only 56% reported to be using Eclipse 4.2 (Juno) and an additional 12.9% using Eclipse 3.8.  The slow down in adoption is most likely the result of the performance issues found in Eclipse 4.2.

At the same time, there was a significant drop in the overall satisfaction with Eclipse. Respondents indicating they were very satisfied or satisfied dropped from 90% in 2012 to 81% in 2013.  Not very good news and hopefully something that will be addressed as the Eclipse 4.x platform continues to mature.

 

Thank you to everyone that participated in the survey. We certainly appreciate you taking the time to provide your feedback.

5 thoughts on “Eclipse Community Survey Results for 2013

  1. “a lot of education and evangelism needed to change developer’s perception of web-based IDEs”

    No, please. Don’t take that path. That kind of attitude will just ensure the demise of the platform. It’s happened before with other products, can’t think of any specific examples right now, but it’s generally better to make a product that people need and want, rather than spend years perplexed trying to convince them they need something they don’t. Listen to your userbase.

    Eclipse 3.1-3.6 was great! Eclipse 3.7 didn’t really add much but it was good. Then everything went to shit. Years later and look at 4.2. Provides no extra benefit to v3, but is slower and has a significantly worse UI. I’ve tried it a few times and I always uninstall it within a few days.

    Look at Intellij IDEA for what Eclipse should be doing right. They have so many more features and that’s what developers want. It’s so sad that JDT et al stopped innovating years back. I really like the feel of Eclipse 3.x and its responsiveness but the feature set has stagnated… and from my perspective the Eclipse guys continue to spend years working on web IDE that no one cares about.

    1. I can not agree more. I especially doubt that there is any value in web-based IDE for Modelling, Embedded, M2M, Desktop and C/C++ folks, who seem to be a substantial part of Eclipse community. However, as far as I know The Eclipse Foundation is only an umbrella organization, and it is not responsible for resource allocation, so more activity in Orion is its own achievement. In any case I would rather focus on attracting more contribution to the Platform and JDT projects, because everything else in Eclipse is either built on top of the Platform or with JDT or both.

      I do not think that 4.2 UI is unusably ugly, but I usually install Jeeeyul Eclipse Themes (https://github.com/jeeeyul/eclipse-themes) to make it more pretty.

      The problems with JDT, as I see it, are in the number of contributors, scope of the work and age of the codebase. Core committers group is small but for historical reasons they have to not only support tooling but also standalone full-featured Java compiler. While Java was stale under Sun there were time to innovate on the tooling side, but after acquisition by Oracle which led to more regular releases of Java most of the work in JDT project is focused on keeping up with new Java Releases. External contributions are almost down to zero, and even those are not always accepted because committers are busy implementing Java 8 (or Java 9 in the future, I predict) https://bugs.eclipse.org/bugs/show_bug.cgi?id=303519 . Only JDT UI keep their site more or less up-to-date. On top of all of that code base is 12 or so years old. In places where I looked it is quite ugly, with classes spanning thousands lines and methods – hundreds.

Comments are closed.