The entire discussion is useful, although I particularly enjoyed Daniel Spock's experiences using project plans for Museum projects since they echo my own frustrations with planning then managing using project software. That's why I so like the flexibility and responsiveness of the Critical Chain and Scrum project management approaches.
Daniel gave me permission to reproduce his comment:
One thing that always fascinates me about these process diagrams is that they reflect a rather poignant human desire to predict the future, to capture the chaos of the unknown and render it in a coherent, logical framework as way to assuage fear. In this way they are almost more talismanic than practical. Unlike statistical charts, they try to project what might be as opposed to what has already transpired.
I work in a museum that manages a great number of complex projects, many of them discrete and overlapping, involving many people and tasks, over a number of years. In my experience, when we use the Gantt charts, they are usually used as a crude forecasting device, an attempt to describe the project in an idealized fashion, as it might happen if all went according to plan.
Invariably, it doesn't turn out that way, however. A vender misses a delivery date, somebody gets sick or leaves for another job, a pipe bursts, the fund raising drive falls short and on and on. (What experienced project manager doesn't have a gut sense of exactly how much "slop" is contained in each strand of the Gannt chart? But you don't spread it around. You hoard it as a contigency, a secret shadow fund of time that doesn't get shared on any chart.)
We could go back and ammend the chart, but this rarely happens. By this time, management is tactical rather than strategic. Because we are all experienced in our respective roles, we are usually already each carrying in our appointment calendars, our PDA's, or our heads what we need to know and, usually, we know know how to pull it out. We respond to change and we adapt. We have many spur of the moment conversations to solve immediate problems, we make decisions, communicate outward and we move on. A forensic ammendation is irrelevant.
The chart, it turns out, scarcely carries the detail, the resolution, necessary to capture these kinds of situational adjustments and their rather more minute ramifications, nor is it very useful to add these back in after the fact. The chart is especially poor at predicting sheer human effort, where it will bulge or go slack. Where the burnout closes in, where someone falls idle with nothing to do.
I think the project chart is a comforting reminder that shows that we know what we are doing, that we have some assessment of all the key variables, a feasible plan. Perhaps its most important function is to act as a bellweather, to trigger anxiety when a blown deadline will pile up onto the next link in the chain. It shows us the time critical points which, when we don't hit them, means we'll have to work especially hard to get things done. It's a confidence builder. But its utility seems to be as a beginning rather than an end.