Submission + - What happened to pair programming? (reddit.com)
ggruschow writes: A few years ago, when Extreme Programming hit the scene, a lot of programming shops were picking and choosing practices they liked and trying them out. What were the results?
Of all the practices, it seems like test-first (at least test-driven development) likely got the best foothold although it's probably most often diluted down to just having a bunch of unit tests (which is still an improvement, although a lot of people let tests atrophy and die when they get desparate). I'm curious how pair programming went. A lot of groups I met were doing pair programming, or intended to soon, but most considered it an experiment. Since then, I've observed fewer and fewer groups doing it. So I'm curious:
Of all the practices, it seems like test-first (at least test-driven development) likely got the best foothold although it's probably most often diluted down to just having a bunch of unit tests (which is still an improvement, although a lot of people let tests atrophy and die when they get desparate). I'm curious how pair programming went. A lot of groups I met were doing pair programming, or intended to soon, but most considered it an experiment. Since then, I've observed fewer and fewer groups doing it. So I'm curious:
- Do my observations match yours?
- How did the pair programming trials you know of go? (Was personal comfort an issue?)
- How is pair programming holding up? Has it become less common?
- How about other XP practices, or XP in general? (I've only seen a couple of groups ever that actually did all the XP practices — Why?).
