Selecting the content that you, our audience wants to see is one of the toughest jobs for any conference. The Program Chair is key to making this work and Mickael Istria, who has served on the program committee for three consecutive years, is no stranger to this. His bio sums it up nicely - he is interested in methodologies and tools to improve productivity and software quality, and to improve a developer's life in general. Here is an excerpt from an interview published in Developpez this month.
What is the goal of EclipseCon France 2015 in your opinion?
In my opinion, the goal of any good conference is first and foremost to give attendees new solutions to their current problems and to help them to discover or create new ideas for the future. This applies to both technical/technological elements that are covered by presentations and technical discussions attendees will be involved in; but also on a human and organisational level, allowing people to detect synergies that can drive them to collaborate.
What are the changes compared to the last edition of EclipseCon France?
A noteworthy change is that we introduced an explicit “Science” track. Indeed, it was identified that Eclipse technologies are widely used in scientific domains, often for data analysis and visualization. So we wanted to give more room to Science use cases and the Science members of the Eclipse community, which covers quite interesting topics that can be useful for many more people than just the scientific community.
We’ll have a hackathon this year, that will have a duration of more than 3 hours, in parallel with the hands-on workshops. During the hackathon, newcomers to the Eclipse community will be able to learn how to contribute to eclipse.org projects: Bugzilla, CLA, Gerrit, SonarQube… All will be explained and demonstrated and we’ll provide attendees with easy bugs to work on with the objective for them submit their first patch to an Eclipse project during the conference.
We’re also proposing a Hacker Space that would be available throughout the whole conference, so that people can easily work together. Sometimes chatting isn’t enough and it’s necessary to spend time sharing a screen and a keyboard.
Finally, we increased the number of (5 minute) Ignite talks in order to cover more topics, and we’ll set up a “Show-Time”, allowing sponsors and some selected projects to run their demos.
Can you describe your role in the Program Committee?
The Program Committee has multiple duties: defining themes that we’d like the conference to cover, making the conference attractive for both presenters and attendees, motivating people to submit presentations, selecting presentations, and also finding new ways to maximize the value of the conference for the attendees.
The Eclipse Foundation nominates part of the Program Committee, and then the nominated members suggest other people to complete it, with the goal to reflect the diversity of the Eclipse Community in the PC composition.
I’m not sure about the formal role of being a Program Committee Chair, but I gave myself the following responsibilities: starting discussions, motivating PC members to be proactive in sharing ideas and opinions, setting up drafts for the various “deliverables” (call for papers announcement, program selection, schedule…) which the PC then used as a baseline for discussions and improvements, arbitrating some decisions when the PC couldn’t manage to reach a consensus (fortunately, this happened very rarely and we tried to use voting whenever possible), taking care of the “management” tasks such as keeping the team focussed on objectives and deadlines, planning meetings, writing minutes… In doing all this, I could lean on the help provided by the Eclipse Foundation staff, who have much experience planning conferences, and by Melanie Bats, PC Chair emeritus, who gave me good bits of advice.
Being the PC Chair has actually been an easy task this year, since other PC members know how to discuss issues efficiently and are respectful of deadlines and everyone’s responsibilities.
How many submissions did you receive?
Choosing proposals must be a very complex process. Which ones where the most debated and what criteria did you use to sort them out?
In our process each PC member votes for each proposal. We then look at the average and standard deviation and debate proposals with the highest standard deviation individually. These are proposals that some PC members love and others dislike, so we try to understand why and we adapt our votes according to the different opinions. Then we focus on the proposals with the highest averages, and we can easily agree to select approximately ¾ of them for the program.
In order to complete the program, we discuss and sort out many talks which have a similar average. Usually, the choice is driven by the following question “given everything that has already been selected, which one of these submissions will bring the most to the conference?”. A presentation on an important topic that is not yet mentioned in the program is prefered to a good submission on a popular but already represented topic.
We’ve created some “personas” in order to represent the audience. This allows us to refer to them and to make sure there is good content for everyone. But in the end, since there were many more good submissions than slots for presentations, it’s frustrating to leave some aside because they’re definitely interesting.
Why should I absolutely go to EclipseCon France?
You have to go to EclipseCon France in order to witness and benefit from the positive entropy that exists in the Eclipse community. See everything that’s done, meet all the stakeholders working around Eclipse, see all the possibilities provided by those technologies and by this ecosystem.
Can we expect surprises?
I don’t think we have any surprises planned at the moment. Our first concern is to make the best conference possible, without necessarily trying to be surprising.
Mickael Istria is the Program Chair for EclipseCon France 2015. He has served on the Program Committee for three consecutive years, with a special interest in methodologies and tools to improve productivity and software quality, and to improve a developer's life in general.
He is an Eclipse RCP/Plugin developer for JBoss, by Red Hat. He has been an Eclipse developer for several open-source RCP projects (JBoss Tools & JBoss Developer Studio, Petals Studio, Bonita Studio - Best Eclipse Modeling Tool 2011, Scarbo), and a committer and contributor on several Eclipse.org projects (SWTBot, GMF Tooling, Nebula, JWT, Tycho, GMF Runtime...). Mickael’s daily work focuses on build, quality and production improvement for JBoss Tools and JBoss Developer Studio, and all related technologies and concepts.