It’s also a bit more than that.
David Allen published Getting Things Done in 2001. The New Economy had just crashed, and the closest thing to an iPhone was still a PalmPilot. Mobile productivity in any useful form was still a good half-decade away. So Allen quite deliberately presented GTD in a low-tech form that everybody could understand: paper. Figuring out how to marry GTD with technology was left as an exercise for the student.
Since then there have been dozens (if not hundreds) of serious attempts to build a digital toolbox around GTD. Most have been garbage. A few have been really nice. But what every approach to the problem has had in common—at least so far as I have seen—is that they are monolithic: everybody wants to be all things GTD to all users.
There are a couple of problems with this approach. First is the question of robustness: whatever the perfect GTD app is, you just know neither Microsoft nor Apple is going to make it. Which leaves pretty good odds that one of those two will eventually kill the unicorn, probably without replacing it. Engineers hate single points of failure for a reason… if my core system might one day blow a gasket without warning, I’d rather lose part of it than all of it.
Secondly… well, what if Application A is really good at Info Management, Application B at Task Management, and C at Time Management? Why should I settle? Why not mash them up and get the best of all worlds?
So that’s why we are here. We spent the past couple of articles analyzing GTD from a functional perspective, trying to get a high-enough level view that we could talk about the process without reference to any technology, whether bits or paper. By the end of the last article, we had finally landed next to something useful: the GTD Shit Cascade.
The Shit Cascade breaks the GTD process down very cleanly into three functional areas: Info, Task, and Time Management. Shit flows in from The World, cascades down the Management ladder, and ultimately either gets shitcanned, gets archived for posterity, or gets done.
Time Management is a bit of a special case. Separate it from the other two, and on its own Time Management is so simple (a calendar) that we don’t really need to dwell on it, so we won’t. Let’s just assume we have it and it works.
The other two modules have very similar structures: one end receives some shit, a process in the middle decides what to do with it, and the other end shovels it out the door. And of course shit won’t walk itself across the street! If we are mashing up different apps, then something has to convey it from The World to the Info Management input, and from that module’s output into Task Management. So those are processes to be accounted for, and if possible automated.
The GTD Shit Cascade is an abstraction: an idealized picture of a generalized process. The next picture is very much more nearly an actual thing:
This is not yet a low-level view of the Machine, but we’re getting there. We can start identifying some specific parts.
I like Evernote for Information Management. Here’s why:
- It is incredibly easy to get stuff into Evernote, and it connects with everything.
- Evernote’s flat structure means organizing it is a lot more like flinging things on piles than it is like maintaining a directory. Flinging things on piles is easy.
- Evernote search is just amazing. There’s nothing it can’t find in those piles.
I like Nirvana for Task Management. Here’s why:
- It is a single-user product: no shared tasks. Shared tasking systems belong at the left end of the Shit Cascade with the rest of The World. Let’s focus on one life at a time, shall we?
- Nirvana implements the GTD model very well, and where it doesn’t…
- … it has a flexible tagging system that closes the gap perfectly.
I like Outlook for Time Management. It just works, plus everybody else in corporate America uses it, too, so using it makes it easier to share calendars, invite people to meetings, etc. But practically anything with a calendar on it will do the trick.
Each is also accessible from every platform you can think of, which is key. Way more detail about these three later.
Remember the metaphor of the Machine as an assembly line? Well, sometimes moving things down the line is a purely mechanical process, and sometimes it requires judgement. The former can be fully automated, and the latter partially… maybe. So on our diagram we have two distinct kinds of processes: automation (the orange ellipses) and manual workflows (the yellow ones).
The Machine is a mashup. That means the output of one stage needs to be connected to the input of another, which is a great spot for automation. The parts requiring judgment (i.e. your attention) need to be as efficient and painless as possible: you want to spend the majority of your time doing shit, not organizing it.
Next time let’s look at the specific manual workflows that live in those yellow ellipses.