As the Agile Manifesto approaches its 10th anniversary, SD Times is speaking to several of its authors to discuss the gathering at Snowbird, what perspective they brought to the meeting, and what they might do differently. The 12 original authors will reunite at the Agile Alliance Conference this August in Salt Lake City.
SD Times: What was your reason for attending the first gathering at Snowbird?
Highsmith: There was an earlier meeting, about nine months before the manifesto meeting, of Extreme Programming proponents at Kent Beck’s in Oregon. Several “outsiders,” including Alistair Cockburn and myself, were invited to that meeting.
The discussions were about promoting XP. I remember walking along the riverbank, talking with Kent and his musing about whether or not “extreme” was too extreme for a brand. I quipped, “What would you call it Kent, ‘Moderate Programming?’ ”
I attended the Snowbird meeting as a follow-up. I knew about Scrum, Dynamic Systems Development Method (DSDM), Feature-Driven Development, and a few other “light” methodologies as they were called at the time. There were enough similarities that I thought it was worth taking a couple of days to discuss our approaches and see what happened. There was very little content planning other than several people wanted to come up with a better name than “light methodology.”
At that point, I’d been working on these issues for nearly 10 years and often felt lost in the wilderness. I figured—and was right—that getting that many somewhat like-minded people together would be a great support system for all my radical ideas.
What area of development had you been working on, and how did you see it meshing with the other efforts going on at that time? Are you still working to advance that specialty? Where is it at today?
I had been working for nearly 10 years on what I was calling Adaptive Software Development, although early on I called it RADical Software Development. I did a bunch or early 1–6-month projects (using 1:1 week-to-month iterations), many of them with colleague and friend Sam Bayer.
I did a project for a major retailer in the mid-1990s. In the previous aborted effort on the project, they spent 18 months doing requirements definition. I helped them do a 6-month project that brought the project back from the brink. At the first iteration feature showcase, the division VP was blown away by having a few stories up and running—especially after the previous fiasco.
In December 1999, about a month after Kent’s “Extreme Programming” book came out, my “Adaptive Software Development: A Collaborative Approach to Managing Complex Systems” was published. The original title of the book was “RADical Software Development,” but I kept thinking something was missing. That missing link turned out to be Complex Adaptive Systems theory, and so I substituted “Adaptive” for “RADical.” Much of that book is about management theory rather than software development. I called the new approach Leadership-Collaboration management as distinct from Command-Control management.
In terms of other efforts at the time, I knew a little about Scrum and DSDM, but didn’t really interact with any of the developers of those approaches. I didn’t know about XP until about a month before Kent’s book came out and we exchanged manuscripts (I wasn’t part of the object-orientation community where many of these ideas had been shared). Mostly I was fighting the entrenched heavyweight methodologies.
There were many project successes, but few organizations went beyond a couple of projects. After a project or three, the organizational antibodies came out and attacked efforts to expand the process to other projects. I’m still working on the organizational changes that will bring agility to the enterprise.
Looking back, is there anything you would have included or struck from the manifesto based on what you know now?
I know there are many people who would add or subtract a few things from the Manifesto, but it has served very well for a decade. I wouldn’t oppose an effort to “modernize” the Manifesto today, but the 2001 version is now part of software development history and should remain as is.
In “Good to Great,” management guru Jim Collins talks about the yin and yang of great companies: preserving the core (values) and stimulating progress (adapting). I think the core values of the Manifesto have been a key reason the agile movement has succeeded beyond my wildest imagination from 10 years ago. As Collins’ research says, we need to preserve the core.
Even those who would update the Manifesto to Manifesto II face the daunting question of “Who [will update it]?” The Agile movement is so big now, how would we ever come up with a group to update it? A credible group? My guess is that most of the original authors aren’t interested.
Where do you see agile development moving in the next five years?
I think the next five years will see three significant trends:
1) Extending the basic agile delivery practices to more fully embrace continuous integration, continuous delivery and DevOps. These will also bring new energy to the technical practices side of Agile.
2) Extending agile project management to other, non-software, projects. The recent involvement of the Project Management Institute into agile is part of this trend.
3) Enterprise Agility. I think we are at a similar tipping point with Enterprise Agility that we were with agile software development 10 years ago. In the last several years, a wide range of organizations—IBM, MIT, Harvard, “The Economist”—have released CEO/CIO-level reports attesting to the changing business conditions and the need for Enterprise Agility.