A few weekends back I participated in the Grails48 hackathon. This was the first time I've worked in a hackaton and it was definitely a worthwhile experience. I've decided to write up a short personal retrospective.

+ Plus

Pairing with someone new

I've written before about what a great experience pair programming has been for me. This was a chance to work with different developers and learn new things from them. I found out about some really useful tools including a neat piece of software for testing: Betamax.

Δ Delta

Lack of planning

We kicked things off without a project. A weekend is a crazy short time to try to create something impressive. Going in with a concrete project and plan would have greatly improved our chance of success. This would have let us think about the problem and explore related API beforehand, and then begin work immediately when the hackathon started.

Lost sight of the goal

Our team failed to submit an entry. While I still found the weekend valuable, this was a bit disappointing. I think the root of the issue was that we didn't keep in mind the deliverables: a pitch video and a deployed piece of software.

If we'd kept in mind the first goal, we would have worked on the pitch video throughout the process. The winning team, for example, had a really neat 24 hour time-lapse video. They also included shorter time-lapse videos of their development effort. The video was a key component and should have received appropriate effort.

With respect to the second goal, we should have started deploying builds much earlier in the process. Deployment issues were one of the hangups that prevented us from completing our project. A first deployment never goes as smoothly as one might think, so early deployment belongs on the list of any project, no matter how small.