I’m a user of the Getting Things Done ¹ methodology for organizing my life and keeping my ADD-addled brain somewhat free from drowning in an inattentive fog. It works, for the most part. The core idea of GTD is that instead of your standard to-do list of giant things like “Do the taxes,” “Finish the Henderson Report,” and “Buy groceries” you break these all down into the individual, specific, concrete “Next Actions,” in the parlance, and do those. So, instead of those big, vague, difficult tasks, it’s a bunch of “Print out the W–2 form,” or “Email Bob in accounting for the Henderson receipts,” or “Get the reusable grocery bags.” Simple, right?

Well, there’s more to it than that, but that’s the gist. Another key part of the GTD methodology is that your Next Actions aren’t put on a big list with all the others. You divvy them up by “Contexts”. A “context” is a specific tool, or location, or even a person, that is required for you to do the task. It does you no good to have “buy milk” staring at you while you’re trying to put together the Henderson Report. You want that on your “Grocery Store” context.

Problem is, with few exceptions like running errands, the whole “context” thing has fallen down in the last fifteen or so years, since Getting Things Done was published. Now we have all these little pocket computers that keep us connected to everything at all times. Some of us work from home, either by choice or by fiat. You could crank out all your @email tasks at your desk, or on the john. The availability of what you need to do your job, or your hobby, or whatever, is no longer limited by geography or connectivity.

This is a huge pain in the keister, if you’re someone like me who does better with hard barriers on their time and attention.

I know I’m not the only one with this problem. A casual Google search brings up (for me) plenty of people thinking about contexts in GTD—thinking of new kinds of contexts that fit how they work, basing contexts on “energy level,” time intervals, streamlining it down to “Active” or “Maybe”, or just plain ditching the whole “context” thing in the first place.

These are all interesting ideas, and I’ve tried many of them at various times. For me, what seems to work is using contexts to divide all my actions and projects between my “@day job” and my “@personal” life. I can then, using my GTD tool of choice, OmniFocus, set up filters and Perspectives to see all my work-related stuff without my personal stuff mixed in, or vice-versa.

It helps, but only so much. I’ve got one of those Knowledge Worker jobs where I spend most of my days in either a text editor, or a web browser. I have a few contexts I use to keep things organized: @online for anything that requires the Internet to get done, @write and @design code for creative tasks, and @email for… email. One the one hand, having all these little subcontexts seems fiddly. On the other hand, if I were to use an @computer contexts for all my actions that use a computer… well, I wouldn’t need much in the way of other contexts.

This is all my manifestation of the problem, of course. Context collapse might manifest differently for you. It might be unwanted texts from the boss and pinging of work emails on your phone while you’re trying to eat a quiet dinner. It might be an inability to work on your side project because the same machine you code on is the same machine you use to play all those cool video games on. Whatever the case, the blurring of lines between our contexts can, without something to keep things neat, turn everything into a morass.

So far, my strategy of a device for work, a device for home, and separate views into the things I want and need to do, have helped a lot. It just feels like there’s more that can be done. As is often the case, I may not have answers, but I do have a lot of things to think about.