To describe the problem, briefly: context switching is what happens when multiple projects and responsibilities get handed to one individual. If you work in the nonprofit sector, you're thinking "Multiple projects and responsibilities being handed to one individual? Never!" ...because if you're been there, you know you're so overworked that it's now beyond a sad joke. And one day you realize that you can't actually account for how you spent the last few hours (or the last few days), because you're mired in a continuous stream (or bog) of attention-grabbing minutiae that seems to effortlessly soak up time, without so much as the voyeuristic joy of, say, Facebook.

I've been reflecting a lot on the evils and art of context switching. It's one of those problems that never seems to go away, and is why successful project management always seems to require a certain level of schizophrenia. If project managers bring order to chaos, why does every successful project manager I know seem to thrive on chaos to some degree? Maybe we were bad clients in past lives, and this is how we get punished?

Generally, we try to tackle this by continuing to try to force order on what is inherently a chaos-driven model (the Gantt charts, obsessive-compulsive scheduling and time-tracking, even more project planning, and client-wrangling), which results in cranky clients, cranky PMs, and cranky developers.

I'm considering a basing project management on relationships rather than on people. This seems like a slight shift in emphasis, but it lets us approach the chaos from a very different perspective. If my job is building, delivering, and maintaining relationships instead of software, then maybe it makes sense that I spend a lot of time on the phone, for example, or a lot of time at events and "social"-ish stuff or even time volunteering with a client, instead of shunning that time to poke at numbers and estimates. Should the numbers and estimates be a smaller part of what I do? Should I focus entirely on building relationships, and trust that projects will therefore be managed effectively? I would be gambling heavily on the value of relationships in that equation.

This is problematic in several extremely unsustainable ways, of course. Clients won't pay for this level of relationship management, and squeezing it into their budgets would seem unethical even if they were willing to. But the more I think about it, the more I think that it would result in better projects. And better people.

I'm going to think about this some more. Maybe the experiment is viable?