While I’m at the XP2010 in Trondheim, I try to update my blog with some of the interesting sessions I attend. This is the write-up from an Open Space session that bothered to think about the tendency to go more and more meta in the Agile movement after all, and whether means that we have nothing really new to talk about-
The session was proposed on the Open Space opening when some people noticed, that there were a whole lot sessions announced on topics like tree hugging and how to be spontaneous. The key question underlying was the question where this more and more meta stuff is going to end up with. Gojko Adzic raised the point, that he wanted to know more about how to develop software better as opposed to how to become a better husband. We had a pleasant talk on this topic over lunch, with a few good insights.
Foremost the discussion arose around Software Craftsmanship and it’s focus back on the technical skills like test-driven development, and refactoring. Gojko also claimed that the focus on how to get testers and programmers work together is very relevant to him, but he is lost on how to learn juggling may help him in any way how to write better code tomorrow. The discussions continued on the topic whether the Agile community has nothing new to learn and teach about, so that it might be time to simply move on.
A tendency was mentioned that many great programmers have left the programming field, and became Agile coaches over the course of the last years. If they are not getting back to the code, this leaves the question where the next great ideas will come from. The discussion turned more interesting, when the group noticed an imbalance in the current community with few advanced practitioner who heard a lot about Agile in the last few years and get bored from it, and the aspiring novices, who need a different aspect form Agile at all.
The discussion got along whether conferences were for more advanced topics, which are not to be found in books for the next one or two years, or on the established practices for the novices. The same imbalance seems to occur in blog entries, not only on conferences, so it seems to be a tendency of the community rather than just the conferences.
Trying to answer the question which purpose it serves to get 30 people into a conference session for two hours to listen to meta-meta-meta-meta-meta-level content, the group remembered the discussions on where to put curly braces that arose ten or so years ago. How much productivity have you been gaining from this? If there’s no more than a one or two percent improvement coming from this meta-stuff, why bother about it at all? This does not mean, that some soft-skills are relevant, but most of the sessions and session proposals just seemed to be misplaced for the only bigger conference that has the name extreme Programming in it.
Of course, all these question may not be answered in such a small group at an Open Space session, but asking these questions on a broader level should surely give some controversy to it, so that we may indeed find out whether or not it’s time to move on.
I have the feeling that a lot of general how-to questions regarding agile and XP have sufficient answers and in most cases the dominant question is: How do we introduce agile to this specific context?
The industry mainstream simply has other questions than we had 10 years ago. In my opinion that is one reason for the shift from programming practices to coaching and for the “community split” that led to two groups of conference attendees with very different interests.
Thanks for your comment, Stefan.
I think we currently go through a value-shift in the adoption process. Deborah Hartmann-Preuss pointed this out as double-loop learning at last year’s XP Days Germany conference, and I applied it to the methodology process overall. Alistair Cockburn pointed out to me that the software craftsmanship movement with the evolution of their manifesto did just that. We revised the Agile values from 2001, took a step backwards, and reflected back on them, in order to evolve the values back in late 2008 out of it.
So, abstracting from the craftsmanship values, we need to get our attention back to practices relevant for good software development. We truly need to build better teams and help them grow. But we must not forget the technical aspects of our work. A team delivering a crappy system and having lots of fun is absolutely of no use. On the other hand a team delivering a great system, but without any team dynamics and fun is probably impossible. Of course, we need to take a closer look on both aspects, the human and the technical, but we should stop to defocus too much. Juggling may be something great to learn, but how is this going to help me write better code or to develop the right system tomorrow? I can’t see the connection here.
On the other hand getting help on how to grow a great team like Yves Hanoulle’s Leadership Game is a very great lesson to take away from a conference. Learning how to lead by getting leadership lessons directly is very relevant to me. Juggling, on the other, not at all.