I enjoyed attending ABC 2009 on Tuesday. One of the things I particularly like about this conference is the emphasis on business, projects, programmes and governance. Too many agile conferences focus mainly on product management and code construction and fail to address the wider environment that development projects operate in.
One of the key characteristics of ABC this year was the number of case studies describing successes (and failures) of relatively large, long-running agile projects. The day opened with Nik Silver’s excellent description of the effects of running a large agile project at the guardian.co.uk which touched on not only the process used and the benefits to the development team and their customers, but also on how the attitude towards technology and the Internet at the Guardian fundamentally changed as a result of adopting an agile approach. This attitude included embracing openness in their content provision – despite the fact that the content is the Guardian’s most valuable asset – and a more collaborative approach to development. Take a look at http://www.guardian.co.uk/help/series/an-abc-of-r2 and http://www.guardian.co.uk/open-platform.
My favourite aspect of the conference, however, was the number of people talking about XP. As I’ve said, Scrum and the like have become the acceptable face of Agile methods and there was a period when the ‘hard core’ of practices necessary for a project to achieve agility was being glossed over. My own personal mission for the conference was to talk about how Pair Programming, TDD, Refactoring and Continuous Integration, combined with good product management and a set of agile project/programme management techniques are all necessary if a business, programme or project is to be ‘truly’ agile. But I wasn’t the only one; several presenters describing their agile projects mentioned Pairing and TDD in particular as key to their success.
Perhaps a down-side to having so many case studies (and as a presenter of a case study I’m partly to blame for this) is a conference that possibly lacked a little in excitement. Hearing lots of people say “this worked for us” or “we tried this and learned something important” is extremely valuable and can be inspiring but it isn’t quite the same as the early days of XP when people were saying “well we’re trying this and we have no idea whether it will work or not … but we’re giving it our best shot”. A notable exception to the sobriety of the sessions was Tim Difford‘s presentation on the use of social media and cloud-based collaboration tools. Although Tim was at pains to point out that his session wasn’t about Agile per se, one of the things that first attracted me to XP was its emphasis on teams collaborating to deliver software rather than an artificial production line of BAs handing over to designers, handing over to coders and so on. I’m excited about how tools like twitter, AudioBoo, Evernote, Balsamiq and our own ChatLoop can help self-organising teams collaborate even when dispersed by time and location.