Following a dzone.com link, I reached an article accompanying a presentation made by Joe Rainsberger at the Agile 2007 conference. This article, "My Greatest Misses: XP 2000-2007", is an honest experience report on the many issues and failures met by Joe Rainsberger as an agile change agent during these years, trying to teach others the merit of the extreme programming approach created by Kent Beck.
Seeing an agile coach admitting publicly its weaknesses dealing with people is an ironic situation that makes it an easy target to criticise agile approaches. If I would have like to add an additional "dark" humour remark, I would have said that the advantage of agile approaches above other processes is that they at least recognise their difficulties to deal with the "people factor", as I never saw this kind of report from a RUP consultant. But I am not so bad. Or am I? ;o)
More seriously, I found that this presentation is a very courageous step from Joe Rainsberger. It is not easy to admit and talk about the problems he met as an agile coach, as it is always difficult for us to recognise our mistakes, without even thinking to make a public presentation about them. Joe Rainsberger does this very sensibly, as he does not blame other people for its own failures, a practice we too often follow. In my opinion, the main message in this article is not about the agile approach, but about changing software development people and organisations. This paper summarises very well the problems we could all face when we try to implement a new approach or a new tool in an organisation. We would like to find enthusiastic people who want to improve their work context and believing in the solutions that we are offering them to achieve this goal. The truth is that we work more in a dilbertian universe as describe in the Waterfall Manifesto, than in an agile paradise. In the reality, there are many reasons for developers to prevent change, to choose other solutions that the one we provide or simply to change at a different (read slower) speed that what we would like them to do it. I am always dubious when people advocates a "big bang" and "all or nothing" change strategy as this has in my opinion many chances to fail (we could have another interesting discussion about the differences between the "appearance of change" and the "reality of change" in an organisation...). There are some situations (very serious crisis, above average people) where change can happens more easily and quickly. However, in medium and large size organisation, you will found that change could only happen gradually and often "imperfectly".
Joe Rainsberg's paper should remind us that it takes many iterations to implement change... and to become an agile coach. Thank you Joe for sharing openly this experience with us, keep on your work and I am looking forward to read in 2014 what you will have learned again in the past seven years.