In the morning we had a nice workshop about the Eclipse E4 platform by Jonas Helming. He explained in details how the new way of implementing RCP application should look like. The concept of dependency injection and application model is really nice, although there is one fatal flaw: this new API can’t be used to develop Eclipse plugins until the Luna distribution will be released. Anyway this was a fine tutorial to understand the basic concepts of E4.
After the workshop and a nice lunch we had the keynote by Bran Fitzpatrick (from Google). He explained how humbleness and generousness can help eliminating the human factor from software engineering. Ok, it is not easy to sum up a keynote within one sentence but I think here “Don’t be a jerk!” can apply.
In the afternoon we took the talk about how can be the development under Eclipse fun again. It was really a mixed bag. The guys showed a lot of bad examples, like the ihateeclipse.com or the lack of the syntax highlight for the unknown java code. On the plus side they proposed a set of preferences and plugins which already make Eclipse better. On top of that a working group was announced to sort out the small but annoying issues from the UI.
The programmers finished with a rather casual meeting with my former supervisor, Istvan Rath. He showed us their new achievement in IncQuery what they are going to show on their presentation on Thursday.
Overall this day was a good start for the conference; I am looking forward for day two.
And now the two of us promote CERN 🙂
I had my first conference talk in Toulouse! It was an amazing to go to the stage tell the people what are we doing. Based on the responses, people liked it too! Overall it worthed the effort we invested in the preparations. But taking part was not the only interesting thing:) When I finally left the venue my head was full of ideas. I would like to note them in this post.
We had a nice introduction to JavaFX. Finally, we have nice GUI solution for Java! I do think, that we can replace Swing with JavaFX for most of the times. I would like to investigate a few things. (1) How does JavaFX handle the usecase where no 3D acceleration available in terms of performance? (2) How can we apply JavaFX on different (Linux/Mobile/Mac/Window) platforms? (3) How can utilize JavaFX in RCP applications?
The second interesting thing was the Tycho workshop. I already have a fair knowledge about the tool so there were no big surprises. Tycho is and will be the number one solution for building Eclipse plugins. I had just one complaint. The descriptors are too verbose and there are a lot of multiplied information (plugin id, feature name, version numbers, etc), which may cause some confusion.
Another interesting idea was to use the Rich Client Platform as a foundation for scientific tools. As it turned out there are a whole lot of scientific tools based on RCP. I think we should rethink how we apply Eclipse at CERN to see whether we could benefit from a possible collaboration with others on this topic.
Mylyn Intent is a cool idea of having our documentation synchronized with the code artifacts. I sure would like to use it to see if I put it in my agile toolbox:)
And so we had the EMF presentation. Well, I was impressed. Both the presentation style and the tool was awesome. The presenter showed how to create an entire collaborative model editing application based on EMF and CDO. After finishing this blog post I will fire up an Eclipse and evaluate the session. If you are also interested, you can reach the resources here (title: Now that I’ve Got a Model – Where’s My Application?).
E4 platform. Hmm… Unfortunately the presenter only showed how can we migrate to the E4 platform but he forgot to say why should we. But still the point was taken; this is the new and better way to develop plugins.
And based on my t-shirt I am officially an HR guy now. 🙂