From the "Extreme Programming" Test Lab
A couple of weeks ago Michael and Babak were finishing off some work for a client. They spent a couple of days working together doing pair programming and blasted through a long list of minor features and bug fixes. To some extent, the "continual code review" made things go very quickly.
This week, we're plowing through a large pile of last minute changes to CityDesk (an original Fog Creek product), and we're not doing pair programming. I have to say that we're moving even faster without pairing up, literally blasting through dozens of fixes and small features a day per person.
Q: To a Political Scientist, what is the singular of the word "data"?
In that spirit, I've gathered some data which shows that the benefits of pair programming are not enough to offset the loss in productivity. We've been making up for it by doing a lot of code review (a process which is made absolutely trivial using CVS/FogBUGZ integration - two clicks from the bug report notification to graphical diffs). Today I spent about 10 minutes total doing code reviews (found some problems, even) which is really cheap compared to pair programming.
You’re reading Joel on Software, stuffed with years and years of completely raving mad articles about software development, managing software teams, designing user interfaces, running successful software companies, and rubber duckies.