That is a trick question.
Despite the fact that there is a blog post once a week or so, and tweets far more often about the nasty work environments people are enduring in the name of Agile, it doesn’t seem to be failing at all. As far as I can tell, agile is still thriving and growing, and probably hasn’t quite reached the point of market saturation.
That seems to be the case at least based on the number of agile themed conferences, meetups, and consultants that are available.
So what gives?
Andy Hunt, one of the signers of the agile manifesto wrote last week about the reasons he considers agile to be a failure. His main reasoning seems to be that agile is based on principles rather than a set of rules. Principles require that people interpret and make judgements based on how things are going right now and how we want things to go in the future. Rules, on the other are simple and less squishy. Thinking and problem solving aren’t required as much for rule based systems. He goes on to tell us about how bad most of us are at thinking and problem solving.
I think that’s probably part of the problem but the reality is deeper. Agile has been commoditized to an almost insane degree over the past 5 years. It has almost become a victim of its own success. Rather than principles, we have SCRUM, test driven development, certifications, kanban and brand name consultants. People differentiate between Agile with a capital ‘A’ and just plain agile for a reason. One of them is a revenue generating brand.
Lots of managers love these canned tools. Programming and testing is complex work, and if you haven’t come from a technical background it can be difficult to understand. People lacking this technical background can get worried by the mysterious workings of their staff.
A lot of the tools I see in agile give superficial measures for how things are going, similar to how the current stage in the waterfall did a while ago. These measures are just as wrong now as they were then, but the comfort they give is real.
The other big thing I notice is that there is a complete lack of focus on individual skill and a reliance on the mythical ‘cross-functional team’. The general idea is that if you get enough variety on a team, you don’t have to worry about individual expertise. Someone somewhere will have at least a vague idea of how something can get done.
Andy’s response to this is the GROWS method. The GROWS acronym stands for Grow Real-world Oriented Working Systems.
To me, that sounds a lot like the agile manifesto emphasis on creating value and getting software out the door. Details are still a little lite, but the most interesting part of the GROWS method seems to be emphasis on individual skill and evidence based development. Andy mentions using the Dreyfus model of skill development which can help to explain how people develop skill from being a complete beginner to an expert. But, unfortunately there is nothing explicit on how that will be done.
As Andy mentions, asking people to think is a tough sell. That seems to be exactly what this alternate his is developing is asking for though.
There are some interesting notes about certification and growth of the method in the FAQ. Mainly that Andy will not be selling certifications, there are some licensing fees involved to teach, and that the method is trademarked. This is all an effort to hold the method close and control how it, um, grows.
Agile is very general and not exactly owned by anyone. So, it grew at the pace of public demand and to the extent that vendors could supply that demand. The way GROWS is described might prevent that, and that might be a good thing.
There isn’t a lot of information available, but I’m interested in seeing how things develop. Who knows, maybe in 5 or 10 or more years people will be writing blog posts like this about the GROWS method.