I have long argued that as a community we are failing dismally when it comes to adoption. I argued once that focusing on test driven development and other technical practices were actually sub-optimisations. Our efforts would be better spent on helping our organisations adopt. For example, why train developers in TDD when you have no customer?

So, imagine my delight when I opened my email a moment ago and found a Doctor Dobbs article, complete with survey, from Scott Ambler claiming that we have indeed crossed the chasm! Scott Ambler is a tireless crusader for improving the software industry and without people like him we’d be doomed. And, all scientific enquiry starts somewhere. So why not here?

What Scott’s results show is that a certain set of individuals and their companies are doing more agile software development than before. The study set had 781 people in it. This is actually quite a small population. What I would like to know is how many of these people where agile fans? Imagine going to your local football ground, Hull City in my case, and asking “what’s the greatest football team in the world”? Answer, Hull City.

The survey was hosted on Jon Erickson's blog. Therefore, I would like to know what type of people read this blog and replied to the survey.

The second thing I’d like to point out is this, 52% of the respondents were developers. Developers love agile software development, and of course, they hate Gantt charts! And surprise, surprise, the poor old Gantt chart comes bottom in the "Effectiveness of Work Products" category. For the record, I am not a fan of the Gantt chart either. However, to some people, it is a useful tool. Do readers honestly think that the results of these questions would be the same if the population was 52% Project Management Institution graduates? CEOs? Customers or representatives from Product Management?

Absolutely no way. Many people I speak to either have no idea what agile software development is or they hate it. They hate it because they can’t make it work in their environment or because they have been burnt with a screwed up implementation. So, in some cases, we can make the very bold claim that agile has crossed the chasm, in others, we can’t.

Based on my observations, agile software development is about as close to crossing the chasm as Wile E. Coyote is with an Acme rocket attached to his back.

In conclusion, it’s very easy to criticise and much harder to contribute. Therefore, here are my ideas. The survey published today, and its counter part from last year, are a great contribution to the larger body of knowledge. They help capture some views and opinions and, as this blog proves, they generate discussion. The population of surveys has a dramatic effect on the results. So, let’s kick on and try to sample larger populations. Maybe we can take Scott’s questions and try them out in our organisations. Then analyse and publish the results as Scott did in today’s DDJ.

Qualitative studies are also very useful. In psychology, large populations are hard to find. Therefore, what tends to happen is case studies precede larger, statistical, studies. As software professionals a great way to share information is with the experience report. When presenting an experience report you always get feedback from the audience. This can help you crystallise your ideas.

Finally, let’s cross the chasm. Let’s refocus on how we can make the jump. What are we doing wrong? What are we doing right? I’d like to see a push towards more education, more on project training, and better dissemination of knowledge. I believe these are areas where we can improve. We also need more of us closer to the top. Us software developers are too nice! We need to work our way up the food chain and start to enlargen our circles of influence.

To Scott and his team, I would say congratulations on a job well done. But I would also ask, regardless of the findings, do you truly believe that we made the jump? In fact, to anyone reading, did we make the jump? Comment below.