Backblog: Anecdotal Evidence
In the next few weeks, I will be going through the backlog of things I accumulated in the last months that I want to touch in this blog. The first one dates back from october, but ties in nicely with my last posts about CUSEC. It’s from Martin Fowler.
This approach - some people report their story, others copy it with or without success - is the backbone of how a profession can learn. The fact that it's anecdotal doesn't stop it from working - after all much of our entire economic system is built on people building on each others' anecdotes of how businesses should be run.
It’s very tough in software engineering to come up with empirical evidence on a particular practice. Some would say it’s impossible. So we have to rely on anecdotes of what worked for someone on a particular project, in a specific context, with individuals, etc. There are a lot of variables involved which make it difficult to reproduce the practice in your situation. However, it’s the best we can get and it doesn’t mean we can’t learn anything. In fact, I would argue we learn more because it actually forces us to acknowledge the differences in what we’re doing and it forces us to better adapt to our circumstances.
It’s important for our profession that people discuss what they’ve tried, what worked for them, what didn’t, etc. This is why I think conferences like CUSEC are very important. It’s a place to gather and share our experiences. The presentation from Motorola on their adoption of agile practices is the best example of this.