L-P Huberdeau

I’m currently sitting in the Apple-sponsored Internet Cafe at CUSEC 2006 (Canadian Undergraduate University Software Engineering Conference). Thankfully, the wireless network is wide open and I don’t have to type from a Mac and an Aqua theme. I will mostly update this entry as the conference advances.

Here is the list of sessions I have attended to:

• Fight the Traffic, by Chad Fowler
• What Hollywood Taught me about Software Engineering, by Robert Sabourin
• Software Testing as a Social Science, by Cem Kaner
• Static Analysis Using the Eclipse Test and Performance Tools Platform

I arrived quite early at the conference. I got to meet other students from all around Canada. When I was heading for this conference in Montreal, I was expecting to meet people from all the universities in Montreal. I was surprised to see there were many students from other universities across Canada. With 250 attendees, the conference is already a success.

The visitor bag contained a lot of corporate publicities. Electronic Arts attempting to recruit, Microsoft logos everywhere, Apple Power Books everywhere. All this explains why the access fees for the conference were so low.

Chad exposed his career path as an introduction, which I would have considered as non-standard if I had never heard of anyone’s career path. Starting as a musician and moving towards computers and software development doesn’t seem that weird. In fact, once you get out of the universities, those who followed a regular path are mostly exceptions.

Since the entire audience is from software engineering and computer science programs, an opening keynote on setting career objectives and looking for the future really was a great choice. From the recommandations: get involved in user groups, projets, open source, blog, meet the speakers and keep aiming higher. I have to agree with most of these, actually, all of them. From my personal experience (short one), user groups and meeting the speakers really help a lot. People like sharing their experiences and helping each other. The amount of information and hints they can give you is a lot more important than what you can learn in school or books.

Robert Sabourin’s talk was mostly entertaining. From popular movies, he grabbed quotes and scenes and used them as methaphors in software engineering to explain common behaviors. While I’m not convinced Austin Powers or Star Wars are the best examples and the content has any value, the talk was refreshing. Since most of the movies discussed were very popular, the crowd really enjoyed it. Sadly, the session started later than expected and they forgot to push back the end, so we ended up missing the last 25%. The talk is probably more appropriate to introduce new students to software engineering than to expose anything to students who have already been studying those elements for years. Still, it was fun.

Cem Kaner’s talk started with a very long list of definitions. The slides contained way too much information and probably could have been skipped since, while not everyone knew those exact definitions, the idea was close enough. He then explained why regression testing was futile in the current development environment, mostly due to the better control over source code. I was glad to hear a university professor saying it. Since the definitions in the begining were too long, the end of the session was once again skipped. Basically, there is no way I can link the content of the presentation to the title. I like the idea of seeing testing as a social science, but I would like to know what he meant about it.

The tutorial on Eclipse static analysis tool didn’t go very far into details, but still demonstrated it could be a tool someone could want to use. At the moment, it’s mostly a framework with a few demo analysis rules. If you intend to do anything with it, you’ll have to write your own rules. It makes sense since each development projects has different priorities, but a larger library could be useful. As a vim user, I don’t really see how it can be useful on a day to day basis, but I will know where to look if I ever need static analysis.

The day ended in a local pub for a party. Something tells me no one here will get much sleep in the next few days.