Friday, December 7, 2007

Does your system work when you have the flu?

GTD Simplicity is so vital to the whole process working.

I was listening to a podcast on davidco and David said two things that struck me about GTD systems:

(1) "these systems needs to work when you have the flu and don't feel like it."

Ain't that the truth. I've finally stuck with a single list manager for months now: Todoist. And I see no signs of it slowing down. The number one source of agreement between me and Todoist is simplicity. I hear and read of so many people looking for a piece of software that will do all the thinking for them. Forget it. If you find it, you've also found that you apparently have no more to add to the world than a computer. In my opinion, the number one attempt at overdoing GTD is in trying to find a way to handle "task dependencies." Which leads me to quote number...

(2) "you don't wanna put call fred if fred's dead i'll call sue...you know that's overplanning... for the most part you just need to know what the kickstart is...that would be really overkill and overplanning to sit down to try to lay out a bunch of stuff that you know is going to be pretty natural and flow as you start moving on action items..."

Yes. Forget dependencies. GTD is absolutely not meant for you to stop thinking about what you have to do. It's geared towards knowledge workers for crying out loud. Perhaps the most important task for any knowledge worker is to be able to use discrimination in deciding what the next step is for moving forward on their projects. This discrimination should be what makes you irreplaceable by technology. Risking being harsh and self-righteous I think attempting task dependencies is one of the most classic novice GTD mistakes out there.

Fine, so if you really want to remember that somewhere down the line you need to do task x but it's not the immediate next action, list it somewhere in a place associated with the project and make sure your weekly review takes care of catching it when it becomes next or close to next. A good place would be a project brainstorm or project steps list distinct from your overall next actions list. If that place has to be your actions list itself, so be it. Just don't worry about dependencies, things change too fast and too much for that to be useful. In Todoist, for example, if I'm mind dumping a bunch of steps for a project I list those as items in that project's list, but my absolute next actions (or MITs if you wanna get your zen on) are ones on which I put a due date for today or overdue so that they show up on my home page. I then work off the homepage until those tasks are done or rendered unnecessary, then I go pick new absolute next actions. This step of thinking after finishing a task (unless it's trivial) is essential to keeping your project management organic enough to allow for creativity, good problem solving, and changing circumstances.