Agile 2009 Reflections and a look ahead at the Agile Movement's Future

Wow, I'm finally catching up after the Agile 2009 hiatus. The main Agile Alliance conference is a wonderful opportunity to share a meal with colleagues whom I only virtually connect with for the rest of the year. While sometimes there are interesting presentations, the agile  conference is a great opportunity to feel the pulse of the agile community.

The pulse I felt is the agile community is quickly maturing from a young child wildly exploring a new, exciting and sometimes dangerous world, to pubescent teenager asking the questions who am i and where do i go from here? All the stuff that I use to do as a child is no longer cute and now I have to learn how to play with the adults. This seems to be creating a lot of confusion and angst for many in the community. First there are many who simply refuse to grow up. One colleague sometimes referred to these people as forming the "Church of Agile" rigidly adhering to the agile old testament and ignoring or dismissing all other practices. I experienced some of this in my own presentation "The Agile Business Analyst" when several members of the audience where indignant that I would even suggest use cases were an effective requirements tool ("that is so yesterday's news!").

Alistair Cockburn set one direction for the conference with his keynote "I come to bury agile, not to praise it" during which he summarized much of the work he has done over the last decade - including his doctoral dissertation. He characterized agile as an ice-berg that dropped into the software development ocean initially creating massive waves  and now diffusing into the main stream consciousness of software developers. I made the date of his presentation - August 25 2009" as officially marking our transition into the post-agile era, when the agile child officially became a teenager.

In seeking an answer to where to next in the post agile era, I couldn't help notice at this conference what seemed like the sudden interest in Kanban. It seems only six months ago, lean was just another agile practice and you never really heard the term Kanban, now everyone is excitedly talking about Kanban, some even refer to Kanban as the Scrum killer. This would be funny if it didn't make me ponder the question is our community one of people who are perpetually seeking simple answers to very difficult questions? Software development is very, very, hard, a wicked problem some say. The beauty of Scrum is that it makes it simple to get started. Growing up is always hard to do. Are we getting into a situation that as agile diffuses into the main stream using Scrum (or your favourite agile method goes here) to take on more complex projects we're beginning to discover that software development is hard? We're discovering isolating the team from change during a sprint is not feasible?  We're discovering stories are not enough? Emergent architecture are sometimes less than adequate? Kanban and lean bring a lot of really good ideas to the table, but Kanban as the Scrum killer? Come on, really! If you're taking that attitude then just wait a couple of years and there will be XYZ the Kanban killer. I hope as a profession we are more sophisticated than that.

Along these lines, one interesting development is the concern some leading members of the agile community expressed about ethics, professionalism, and isolationism. During an informal lunch, none of us could quite put our finger on what we felt was wrong though the issue of Scrum certification became a bit of a lightening rod for this concern. Beyond just the Scrum certification issue there were concerns about our professional ethics, sometimes characterized by self proclaimed "coaches" underbidding on projects. There were concerns of the insular nature of the Agile community, that many of us believe the world should come to us, rather than us reaching out and learning from other professional colleagues. These kind of questions and this kind of introspection is healthy because it demonstrates a desire to grow.

Many of the sessions as usual, used cute and entertaining games to make their point. One by Nate Oster and Abby Fitchner "Where does developer testing end, and tester testing begin?" that had many cute question and answer puzzles to demonstrate the value of collaborative work between developers and testers. Nate and Abby are very entertaining and the session was well run and the points well made. I even was pressed gang into serving as a judge with Dan Rawsthorne for Rod Claar''s and Doug Shrimps "May the forces be with you, Exploring the forces Driving and Restraining Agile" While sessions like these avoid the usual dreary drawbacks of the "death by powerpoint" type talks are we getting an effective message out? 

However, if you are anything beyond an agile beginner this kind of session really lacks a lot of solid take away value. Yes, I know that getting people to work together is important, tell me how do I do that? How did you do it? Do you have lessons learned? Good patterns? Can you give me guidance for how I may apply these ideas in my situation? Where are the new ideas? This is a concern I had with a lot of the sessions, there were many, many introductory sessions, some which are about agile basics and have not changed in years. I expect the premier agile conference to be a place where the new ideas are transmitted.

Now after that rant, there were a few good ideas put forward.

Linda Rising who always has some new insight on agile in her presentation "Agile: Placebo or real solution" asked the question "does agile work because we believe it will work" How do you answer a question like this? Does it matter? Bravo Linda for once again asking us to think.

Dan Rawsthorne raised the agile bar a notch with his agile metrics presentation. Dan has been working hard to craft useful metrics for agile projects beyond the sprint burndown chart.

David Anderson introduce the idea of using real-options as a model for software projects. While the math behind a real-option is overwhelming, even just thinking about backlog features in terms of real-options may have value in managing risk, and communicating that risk to our stake holders.

Speaking of risk, Chris Matt and Todd Little described their experiences and ideas for managing risk. What concerned me was the small number of people who attended this session (Chris and Todd are well known and good speakers) and that most who attended the session were definitely in the 40+ set. Have I missed something, is risk management no longer an important issue?

The topic of risk management was raised again in a well attended presentation by Tim Hughs and J. Jenks "Adopting Agile in and FDA Regulated Environment". The bottom line, the FDA regulation is not an impediment to agile adoption, so long as you dot your i's and cross your t's like you always have. Perhaps this should have been called "the agile paper trail"

My conclusion, the agile community has reached or is reaching a fork in the road and is looking for direction on how to proceed. Growth is a scary thing, and just like that pubescent teenager asking who I am the agile community is trying to identify and differentiate itself as our ideas melt into the ocean like Alistair's proverbial ice berg. I believe what scares many in the agile community is that like many rebellious teenagers we are beginning to discover that our parents are not as dumb as we thought they were, and actually have some pretty good ideas about how to do things.