I am far from advanced or stable in my GTD implementation, so slather
anything I say in this regard with a heavy helping of caveat. However,
here are a couple of discoveries I've made that may be helpful to
others:
1. The only times I've been truly successful in plowing through a
mega-task such as "revise manuscript" in anything close to a setting
have been when what "revise manuscript" *means* have been made quite
explicit. I guess I could say I do and don't make NAs for this - I
don't in the sense that I rarely add this to my regular NA lists (which
I keep on the Palm) unless they're very very few (otherwise I find my
lists get too long and unwieldy and overwhelming and I have difficulty
self-managing under those circumstances). I *do* in the sense that
I'll make a section-by-section outline (I'm primarily windows-based
right now, and often use Ecco Pro for this) of the
article/paper/report/whatever and detail out what's between the
paper-as-is and the paper-as-it-needs-to-be for its next step on the
journey (submission, resubmission, review by chair or another
colleague, etc.) Some of these will be fairly complete NAs in and of
themselves - e.g., "Tighten up this paragraph," "add more cites on
___." Others will prompt for me things intermediary steps that need to
happen - get more info on X, decide about Y, ask Z for input on how
well this frame works, etc.
2. Despite what that sounds like, I'm not as meticulous as I could or
should be about breaking everything I do down into micro-NAs. I do get
there when I realize that I'm procrastinating or having difficulty
getting my head around what I need to be doing. Otherwise, unless this
need presents, I tend to break things down into slightly chunkier
levels. Informally, I think I have a model in my head similar to the
lingo used by a dim sum restaurant I used to frequent - they organized
things by "nibbles," "small eats," and "big eats." I don't try to
avoid "big eat"-sized tasks entirely, because I know there's some
things it helps to just give myself space to immerse into the thought
process of the task. But recognizing when things *can* be considered a
nibble or snack is good, too - then I can play games with myself, such
as alternating between big eats and small eats, setting a timer for
knocking out nibbles, etc.
--Shelby