Conclusion

Our experiment with mob programming, still in its infancy, has so far received mixed reviews. The undeniable benefits of the practice have been more indirect. Communication has been noticeably enhanced. Test coverage is improving. The meetings themselves are becoming more productive, regardless of format. There is a growing sense of ownership of the meetings by the team as a whole. We view these meetings as a work in progress. We do not view the process as it exists today dogmatically, but as a lightweight method of encouraging shared values. If mob programming does not work, we will try something else. We hope, however, that sharing our experiences may help others searching for ways to facilitate the transition to a more people-centered, communication-focused process like XP.