CUSEC 2006: Day 2/3
The second day of CUSEC started with only a few people. Most of the attendees decided to get a few more hours of sleep after yesterday’s party. They probably should have selected to coffee option, because they definetly missed a great session at 9. Here is the session lineup for the day:
- Considering AJAX, by Chris Laffra
- Modular Concurrency, by Peter Grogono
- Testing in a Creative Environment, by Karine Roy
- A Panacea or Academic Poppycock: Formal Methods Revisited, by Connie Heitmeyer
- Model Based Development of Advanced User Interface, by Peter Forbig
- Creating Passionate Users, by Kathy Sierra
As a side note, this entry has been written in multiple parts, so don’t bother if things don’t really fit together.
The AJAX session covered most of the debates around this new emerging technology. Quite a lot of application examples have been presented. Most of the arguments given were nothing new to anyone who followed the evolution of the technology in the last few years. Still, the examples demonstrated a very nice future for rich application development in the browser. Writing an AJAX framework is a total waste of time, simply because there are so many of them available already and there are too many things to consider. It’s not only about interaction between the client and the server. According to me, what’s missing right now is frameworks that have a fall back plan for browsers with no JavaScript support and other various accessibility requirements. Unless the work is done twice, there is no real way to benefit from the advantages of AJAX without leaving behind a minority of users. Doing the work twice is always an option, but a very expensive one.
Modular Concurrency was really the kind of session I was expecting here at CUSEC: A session that challenged the current technologies and proposed a vision for the future, without taking down everything that was built before. The basis of the talk was that the current languages are poorly adapted to the context deploying execution across multiple processes and did not match the higher level models we build. The origin of these problems would be that the evolution was made by finding direct solutions to the problem instead of finding the actual cause of the problem. I don’t think I have half the knowledge required to give an accurate explanation. I will probably listen to the talk once again once the podcasts are available (I’ll try to remember to link it from here as well).
One of the very good parts was that the author started by presenting many of those giants
who built software engineering as it is today. At the time, the technology was a restriction, but they already had the solutions written down. What I really liked is that the speaker convinced me that there would be major changes in the way things will be done in the next decades.
During Karine Roy’s presentation, there was absolutly no doubt it was about Autodesk recruiting new talents. The topic was “what is a tester and what it does at Autodesk”. It was quite interesting to see all the testers actually do, but when it comes to practical uses, this session had no real content. I wasn’t expecting much more about of a corporate presentation. The other option available was a presentation on Eclipse, so I think I still made the right choice. It was a good introduction to the activity period now, which is a career fair where people run around with CVs for jobs at Microsoft and gaming companies. Mostly amusing to watch.
Formal methods are a totally different universe. These talks really remind you this. It’s facinating to see how much they can do with those languages. The only problem is that before writing the format requirements, you actually need to get the right requirements. If you don’t get the right ones, you’re not verifying hte right stuff. The talk didn’t contain enough details to really see how it’s done and how much it can help finding the real requirements, but it can be excused due to the very short duration. I definetly want to dig into this in the future, I only wish I will be able to find a job where I can apply it.
The fun part about people from the millitary industry is that they always begin their presentations with the exact same examples and data. I have no idea how many times I have been presented the Standish group’s CHAOS report. Don’t these people realise we already heard about it? No need to spend 15 minutes explaining that too many projects fail. We know it. So much more content could be covered without this waste of time.
The last session of the day was definetly one of the most impressive sessions I have ever seen. Creating Passionate Users was not only a presentation, it was a live demonstration. The idea is nothing really new: focus on the user and do everything possible to get them to enjoy their experience and appreciate what they do. What was great is the approach to the topic. A lot of the content was about the human mind and how we can attract the attention on the content by adding other various elements. I don’t know how much of this I will be able to apply in my day to day job, but I will definetly put some effort into it and try to keep this spirit.
The day ended at the restaurant, where there were no places available when we arrived, even if we had the entire place reserved. The service was horrible, but we could still manage to have some fun. Other than this, the day was quite important for me as I decided to get involved in the next edition of the conference in 2007. When John Kopanas first asked me yesterday, I wasn’t too sure due to my very tight scheddule, but it’s a very nice project and everyone in the team is very dedicated. It’s nice to see how all the universities get together for the event.