Working Up

by Dwayne Phillips

A recent discussion with Johanna Rothman reminded me of terrible discussions I had with government project managers about project lifecycles. The shame is that most government managers don’t know one project lifecycle from another. Hence, when talking to them, ask enough questions to understand what they think they really mean. In general on almost any topic, read back in time until finding the seminal paper.

I have been discussing project lifecycle models with Johanna Rothman. I reviewed a manuscript of her upcoming book on managing project portfolios. Our discussion has been about interim deliverables and such that help projects with a serial lifecycle be more like iterative projects.

Our discussion has reminded me of a painful part of working in government with government managers:

Most government managers don’t know one project lifecycle from another.

This became obvious to me when hearing government managers mention spiral development. Barry Boehm created the term spiral development. Boehm had specific ideas in mind with this. The core is about reducing risk through spirals. In each spiral is a point where you decide to continue the project or stop work on it right there (a.k.a. “kill it”).

Well, look at most any mention of spirals with Department of Defense or other government publications. These confuse “spiral” with anything where there is more than one point where delivery occurs. Gasp. No wonder many government managers are confused. Most of them stop reading when they read one thing. They don’t bother to check other sources or read older documents that use the same term.

I guess this pushes me to recommend several things: First,

When a government manager mentions a project lifecycle, ask plenty of questions so that you understand what they mean.

Second:

Read back in time until you find the seminal report, paper, or speech.

If you want to know what spiral development is, read Boehm’s original paper on the topic (a reference to it is found here). This holds true for any topic. If you read something, read the references. Keep reading back through the references until you have read the original. It is surprising how people can take the name of a topic and twist it around until it is a big mess. This is true of the original “waterfall model.” By the way, this Wikipedia article on the waterfall model shows the most-often shown diagram, which is not what the original paper recommended. The original paper showed many feedback paths that correct the supposed problems with the model.