I wish to continue this series with what is to me the most provocative and interesting new directive for project management that I have seen in a number of years: setting up the ability to "trim the tail" on a project in order to choose between early (or timely!) delivery and later (but more complete) delivery.
I refer to this as a provocative and interesting "new" directive because I have rarely seen it done deliberately, it gives some quite different advice from standard PMI or standard agile wares--and it opens the door for using agile methods profitably on large and complex projects.
The strategy consists of three parts:
Name known technical and social risks as early as possible (in theory you are doing this anyway), and address them with running, tested features in the smallest possible way as early as possible in the development period.
Identify and prioritize the highest business value items to the front of the project, and implement them with running, tested features in the simplest still acceptable way next.
Defer quality and feature improvements to the back of the work queue in such a way that at any time the value of the system is highest for the time spent.
In the first period, the team constructs a "walking skeleton" of the system, implementing some
"Technology is a gift of God. After the gift of life it is perhaps the greatest of God's gifts. It is the mother of civilizations, of arts and of sciences."