New Study of Eclipse User Community

We have recently completed the second annual survey of the Eclipse user community. This is a study conducted by Evans Data and is sponsored by the Eclipse Foundation and some member companies. Although most of the study is only available to the sponsored companies, I have posted a small subset on the wiki.

Some of the highlights from the survey include:

1. A confirmation that RCP is really gaining adoption and awareness. 22.7% of the respondents are using RCP; more impressive is that 68% intend to use RCP in the next six months. To me this is great news! We are issuing a press release on Tuesday to announce the fact.

2. We also asked a question to determine which Eclipse projects were being used or planned to be use. No surprise that JDT came out on top with 57.5% but CDT came in second with 47.3%. I think this is amazing and a big congratulations to Doug and the CDT team. Could there be a day when more people are actually using CDT than JDT? It is also great to see EMF being used by 36.5% of respondents. I still believe EMF is one of the hidden success stories in the Eclipse community.

One stat that did surprised me was that only 29% were using WTP. I know the team is executing really well and I would expect they will see a big improvement next year. It could also be a fact that a lot of developers are using the commercial equivalent of WTP?

3. We also asked what are the barriers to adopting Eclipse. The number one barrier was ‘Lack of Training for Eclipse’’. This was also a key issue last year. Indeed, when I am talking to people about RCP, ‘where can I get Eclipse training?’ is often one of the first questions. I think Eclipse training and services represent a great opportunity for organizations. Organizations like AvantSoft and RCP Training Alliance are leading the charge but I am hopeful that we will see more companies offering Eclipse and RCP training.

4. I must say it surprised me that over 80% of the respondents use Windows as their development platform and Linux accounted for less than 10%. However, when we look at our download numbers this is pretty consistent. Eclipse users are using Windows for their development platform. Deployment is slightly different, with Windows just under 60% and Linux around 20%.

In general I was very pleased with the results. These types of surveys are one measure of how well the community is doing and by this measure things seem to be pretty darn good. 🙂

4 thoughts on “New Study of Eclipse User Community

  1. Congratulations to everyone at and behind Eclipse — this is certainly great news.

    I think the point regarding the WTP is probably fairly likely; the commercial derivatives certainly use the J2EE extras as a selling point. It would have been interesting to find out whether the superset of commerical products using WTP were included.

    I think that the CDT is great, but it will be really interesting once it supports C#. That’s probably one of the last things that it will get though (after all, there’s not many other C-based languages, unless you count Objective-C — and as much as I’d like that to be in there, I can imagine that there’s not much demand for it 🙂

    Anyway, good job!

  2. I guess, WTP will be used more, as soon as it is less buggy. We are all using it here, because it is great in many ways, but it is, still, the most buggy ‘standard’ Eclipse plugin, by far. Complete Eclipse crashes occur quite often when using WTP. I’m looking forward to its next bugfix release!

  3. >> “…what are the barriers to adopting Eclipse. The number one barrier was ‘Lack of Training for Eclipse’. This was also a key issue last year.”

    Eclipse training is definitely a niche where corporate partners can avail themselves of Eclipse’s complexity to generate more revenue. Most users I interact with suspect they don’t know the half of what it can do (or choose not to do it because the process is too cumbersome for them to bother). Training is really a must for a developer to be productive in Eclipse, given the wide range of plug-ins.

    But doesn’t this suggest a more fundamental problem affecting Eclipse’s adoption? Developers, more than people of other professions, are technically savvy self-learners. If the IDE is so complex that it becomes a barrier to its adoption, maybe the complexity itself is the problem, not the lack of training. I think too often we equate complexity with robustness.

    Don’t get me wrong, it’s great that Eclipse has such wide vendor support. But how do we rein it all in to make it a unified and (at the risk of sounding redundant) integrated IDE? Like a coworker of mine said recently, “Eclipse is really powerful, but sometimes I feel like I’m using several IDEs instead of one.” Perhaps that perception could be construed as Eclipse’s strength. More often than not, however, it’s cited as its weakness–and one that is rather unique to Eclipse in the broad Java IDE market.

    Still, adoption of Eclipse means adoption of, or continuing investment toward, Java technologies, and that’s what’s most exciting to me.

Comments are closed.