In my line of work we build and sell expensive instruments for analyical chemists that are driven by large software data systems (3-4 million lines of code). These systems have to control and optimize the instrument, store and analyze the data for 30 different applications, enforce security policies for provenance of the data, enable distributed data processing, and produce all manner of different printed reports. When my manager says to me “How long will it take us to move all of our existing functionality from the Mac to the PC and what will it cost?” all I can say to him is that I need to convene a scrum team of 7 people and we will self-organize into a team that will develop an empirical process which will iterate toward an estimate. While that may be the most honest answer, you’re not likely going to be asked to do the job. It would be far better to be able to say “This is the process by which one scrum team managed to deploy scrum across a group of 28 developers and delivered a product of similar magnitude in 3 years. If you’ll allow me to form a team, I’m sure we can empirically create a process process that will enable us to give you an estimate.”

I’m a PMP and have to manage different projects using different ’styles’ based upon the need, complexity, and product to be delivered. To assume that, as a PMP, I have all the tools necessary to ‘develop/build everything’ would be proof that I wasn’t paying attention when I was learning about PM. As someone said in an earlier post, PMI does not deal with how the work is to be done, but in how to manage the process to ensure a greater possibility of success. How the work is to be done is really dependent upon the nature of the project and the team. Just as I wouldn’t use every tool in my toolbox to drive a wood screw, every tool/process in the PMBOX is not appropriate for all projects.

In our development efforts we’re attempting to use more ‘agile’ methodologies while working in an environment (government) that wants to see everything planned out prior to releasing funding and before work can begin. Since we’re spending other peoples money there’s wisdom in the approach but only to a point. My current view is that agile methodologies can be applied most effectively in the execution and control portions of a project where the actual work is accomplished while the appropriate PMI methodologies for the project can be effectively applied in the initiating, planning, and closeout portions of the project.

This approach may be overly simplistic to many but I’m open to learn. Thanks for the time.