Friday, December 18, 2009

The Mythical Man-Month meets the Yellowverse

In 1975 Fred Brooks wrote a book that he called 'The Mythical Man-Month'. In that book he claimed that assigning more programmers to a project running behind schedule will make it even later, due to the time required for the new programmers to learn about the project, as well as the increased communication overhead. I suggest that with the introduction of corporate social software Fred's theory is no longer automatically applicable to the world of software development, because when I want to add a new programming resource to a project I don't always need to involve a new programmer.

Let me explain...

Serious social software tools (which excludes Facebook) foster a community of software resources who are willing to add value to your project without being formally involved. They post code samples and tutorials on their blogs (thanks Declan et. al.) and respond to technical questions in software forums. We all have different reasons for doing this but the end result is that our skill set is made available to other Lotoids without us needing to be fully briefed (or even aware of) their project. So a Man-Month of work for a clued-up developer can be made more productive on demand by incrementally adding the diverse skills of the Yellowverse as required.

The other side of the Man-Month is very real financial pressure felt by larger IT organizations who have a need to optimize their resource usage. If they are paying a developer $100k per year then they naturally want that person to be billable for as many days as possible with minimal management intervention, and the easiest way to do that is to put them on a customer's site for a large slab of time - three months is good and six months is better. And if they can find a way to justify a Team Leader and another Software Specialist in the project then that makes even more profit for them.

In my experience the smaller Business Partners don't work like that. When you are a team of only three or four consultants it doesn't make sense to tie up one third of your resources for six months at a time. You want to get in there and finish the work ASAP to free yourself up for the next opportunity.

So if we look at a paying customer who has a half finished spec (hey... he's a Retailer, not a Business Analyst) then how does he get maximum value for his money? Is his need best met by an organization whose business model thrives on throwing more bodies at the problem (thereby raising the total cost and invoking Fred's curse on his project), or does he get better value by engaging a Business Partner who knows how to work the Yellowverse and how to tap into the zero-cost skill sets of the Lotus Legends?

The answer is obvious... the problem is finding a way to say that into your proposals to customers.

1 comment:

Mary Beth said...

I found this a really thought- provoking post, and I'd like folks to supply some of their own real-world examples if they can!