As I said yesterday, I am currently reading Algorithms to Live By: The Computer Science of Human Decisions Book by Brian Christian and Tom Griffiths and in particular the section on scheduling.
The thing is am finding really interesting about this section is that it helps explain why I find techniques such as GTD1 and Inbox Zero so compelling and effective. The book explains about the cost of context switching.
In computers there is a phenomenon called thrashing, where after getting asked to so one more task, the performance completely falls off. Like a juggler getting asked to juggle just one more ball, leading to all balls getting dropped. This is due to the computer spending more time on meta processing, than doing the actual processing. This includes, working out what the highest priority task is, what resources are required, transfering any missing data from storage among other things.
People have the same problems. GTD and Inbox Zero help by simplifying the decision process around which task to work on next. If you have lots of emails in your inbox, it can take a while to work out what is the most important to work on first. Inbox Zero says that you will process all emails in one go so there is no decision required, do whatever one you click first. You may not respond immediately, but you will work out what needs to get done, and put that task in the appropriate Todo list, so that they can be prioritised later. GTD helps prioritise the Todo lists. It hides tasks which cannot be done at that time, so they don't get in the way.
When you don't have to waste time working out what you have to do, then you can be more effective with the time you have.
I also liked the comment about getting out of the problem of being completely overloaded that you can't work out what is most important. Give up trying to work it out. Just do the first task you come across, then the next. It doesn't matter which you do, just do some. That way you can make some progress and not waste any time with the meta work..
- Getting Things Done ↩