Organizing past and future research

I'm having trouble splitting my attention between two projects, PhD research for AGU in December and the project with geology. I've always been good at organizing things, but it's more difficult when the goals are hazy and the scope seems overwhelming.

J-man likes to clean by a method he calls "convergent messes." He sorts things into piles based on where they go. First maybe it's by floor, then by room, then by place in the room. Eventually everything is put back where it goes. Planning my PhD should be similar. First I define specific, detailed research questions. Then I decide the experiments I'll use to answer them. Or do I choose data sources? When do I outline the papers I'd like to publish based on my work? I feel like I'm flailing after defining the questions, simultaneously trying to plan everything else without being sure exactly what I need to plan. It's the prelim again and again. So far my plan is to start with what I turned in for that and refine, even though revisions are not required since I passed. At least it provides a starting point.

On top of planning, I also need to record what I do as I do it so I don't get lost along the way and make writing my dissertation an epic nightmare. I've never been good at keeping records. I don't remember if I ever even balanced my checkbook (though I know how). I record the checks I write only because I don't have duplicate checks.

I've skimmed several sites that tell researchers how to keep a useful notebook, but they don't seem to address some of my questions. What do I write about research as I do it when my work is mainly programming and computer modeling? Surely I don't need to document every step of debugging a program, but I'm sure I need to record when we make a major change, such as adding random noise to the initial conditions of a simulation, and why we change it. Should I keep a separate notebook for each project or one master notebook?

I asked my advisor about keeping records and he relies on comments in code and TimeMachine backups. He doesn't seem to keep general notes in one place. He also often seems disorganized. I suppose I'll take a hack at it and make up the rest as I go.

Yeah, that usually works out.