Anders Lybeckers Weblog!

The third and last day at the JAOO conference was the best in regards of the sessions I attended. My day session schedule looked like this:

• “MonoRail: Building Maintainable and Testable Web Applications” by Oren Eini and Hamilton Verissimo
• “Fault Tolerance” by Robert S. Hanmer
• “Performance Art” by Kevlin Henney
• “Applying Craftsmanship” by Pete McBreen
• Panel discussion: “Enterprise Application Frameworks” with Graeme Rocher, Eberhard Wolff, Erik Dörnenburg, Oren Eini, Justin Gehtland, and Alef Arendsen

I have written an abstract of the session “Applying Craftsmanship” by Pete McBreen in a separate blog post, because it deserved it. It made me think (which is a good thing ). Read it here.

JAOO 2007 conference summary
I enjoyed the JAOO conference, but was a bit disappointed with the technical level. Again, maybe my expectations were too high? But then again, I will consider coming back next year. That has absolutely nothing to do with the fact that I won an Apple iPod at an IBM contest.

Below is a picture of the price I won in the IBM contest.

The session “Applying Craftsmanship” at the JAOO conference by Pete McBreen was by no means technical but made me think. The stuff he talked about is, in my opinion, common sense. Below is an abstract of what I got out of the session. I wanted to share it with you and wrote it down so that I don’t forget about it in 6 months time.

• Currently the norm for gaining the title as senior developer is only 5 years of experience. Does that make sense? Can you really gain enough knowledge in such a short time? What is the next step in the career path for technical for a developer? Architect or project manager? Not all developers make good architects, as the skill required differs. Almost no developers make good project managers. What is often seen is the organization ends up with a lousy project manager and loosing their prime developer.

• Why have two teams – one to develop the solution and another to maintain it. It does not make any sense, as the initial developers are the best source of knowledge of implementation details. Give the initial developer’s incentives to build the best possible solution by having them maintain the code base afterwards.

• Developers fresh out of college or university should learn from the more experienced developers (senior developers?). The best way to learn is by looking at others code or paring up with a more experienced developer. Pete McBreen favored having the newbies help maintain existing code. That makes sense in my ears, as they then get exposed to others (hopefully) high quality code, while paring up with the experienced developers.

• Another issue he addressed was the current trend of discarding existing code and redeveloping the same functionality over and over again. He made a point out of continuously improving the software quality, so the need to discard the existing code would be pointless. The goal is not solely to continuously refactor, but to improve the user experience – functionality wise.

• This ties very well to the last point of shifting the developers’ focus to the system functionality that improves the users’ experience. The users are paying for our salaries so of cause we should focus on their needs. This requires that developers talk to users and understand their needs, because developers are not domain experts and do not have any clue of the users needs.

Pete McBreen has written a book about the subject called “Software Craftsmanship: The New Imperative“. I haven’t read it, so I can’t tell you if it is any good. Maybe I should read it…

I am on my way to the JAOO conference in Aarhus, Denmark with a colleague. We booked a late train and ended up on a geek train!
Addicted as we are, we logged on the Internet via our mobil phones and noticed that we were not the only ones with Bluetooth enabled laptops/mobil phones!
This was what we say.

That is pretty amazing as Bluetooth is a personal network and the range is about 10 meters.