Planning the Plan relates to the planning process itself and not the concrete deliverable. It is the entire team’s responsibility to participate in the process, however, during the inception phase of the project, the responsibility for Breaking Ground falls most squarely on the shoulders of the Director.
The Producer at this point has tentatively decided to pursue the project and has some vague idea about the details of the value proposition. It is up to the Director and the team (management, sales, marketing, technology, etc) to clarify the value proposition and devise a strategy for attaining it. In order to expedite the process the Director must frame the problem by developing a “straw man”. The straw man contains the Director’s initial thoughts related to the following non-exhaustive list, depending on the project:
Once the Director has gathered and (crudely) documented his or her initial thoughts he or she meets informally with key stakeholders to surface the ideas and refine them. The straw man is a device that helps plan the plan. It is pure tradecraft—far removed from anything that vaguely approximates a science, anyone that claims otherwise is completely (and utterly) full of shit.
The straw man should take no more than a couple of weeks to develop. Once developed it is time to meet with the entire team for a kickoff discussion. The purpose of the kickoff discussion is to interrogate the straw man. The interrogation(s) results in a list of tasks required to fill the holes (i.e. identify risks, gather additional requirements, develop rigorous use cases, further define project economics, etc., etc.). Filling the holes should take (approximately) 2-8 weeks, depending on resources, complexity, etc. If the holes cannot be filled to the team’s satisfaction, then throw the fucker out and start again.
At the point that holes have been filled the inception phase of the project is complete. Do you have all the answers that you need? Can pigs fly? The RUP defines only four project phases:
The Elaboration phase contains iterations that are aimed at reducing risks and getting real answers. OK, so what’s the point? It is an exercise in futility to drag the Inception phase on and on in search of perfect requirements (use cases, economics, or any other fucking thing). Software is n-dimensional, temporal, and organic, a living and breathing beast.
You will never be able to model the beast on paper. If you want to understand the beast then you need to interact with it! Only by interacting with it can you determine if the beast should live or die. This determination is made during Elaboration.
Elaboration iterations should be planned in the same way. Develop an iteration straw man. Get some feedback. Refine it. Meet with the team and get task estimates. Refine them. The planning process should be as iterative as the development process. Expedite it by Breaking Ground early and often.
If you make it to the construction phase then, by definition, you have all the answers you need to give the beast life (or so you think). During Construction you build the beast. During Transition you cut the umbilical chord and hope that the world treats this thing that you have hatched kindly. Hopefully it survives birth, adolescence, college and goes on to have a long and distinguished career.
Planning is a motion creation tool. Planning reduces entropy by organizing chaos. A plan is outdated and wrong, with respect to specifics, as soon as it is committed to paper, however without one you cannot proceed. It is a development tool and nothing more.
Learn to plan the plan and don’t hesitate to modify it when necessary. The actors must help write the script.