Went on a one-day course in agile development today. Truth to be told I didn't have any expectations of learning anything. On an individual level, I've always worked agile - i.e. I get at least a minimum done and try to spend as little time as I can on administration. But to my surprise we were treated like adults during the session, and I learned a new thing: how pressure can enforce quality.
See, Volvo has this idea of using one-week sprints as part of their recently-incorporated agile development programme. Previously I dismissed one-week sprints as nonsense; we'd spend more time on starting/stopping, administration, deployment, database schema delta propagation and technical hick-ups than we would on actual development during a single week. And I'm sure we would - in the current state of affairs - but the point our teacher made was that only a project run with high quality (both in terms of backlog, code and automatic tests) can cope with, and get something done in, one-week sprints. I now believe he was right.
What remains to be seen is if we can improve our quality enough so that the team gets things done in a week. That is a trickier question, with a much more tiresome answer.
I've finished the porting+tweaking of Push to iOS a few days back. In a few more days of rest I'll add a tank-style controlled hovercraft with shooting abilities to Push for prototyping and I'll try some multiplayer to just get the feel for the computer/iPhone interaction.
