A draft from my personal-productivity journal.


A natural 2x2 falls out when you organize tasks by how clear you are on their start/end times (or the amount of time they'll actually take), and how clear it is in a non-time sense that you've started or finished it. I call these two axes time and task delimitation.

Time delimitation and task delimitation defined

Time delimited

Things which are highly time-delimited have clear, natural start times and end times. Things can also be considered highly time-delimited if you know to a good deal of accuracy about how long they will take. Things which are less time-delimited are fuzzy as to when they start and end.

Task delimited

Things which are highly task-delimited have clear, natural non-temporal start and end states. Things which are not highly task-delimited are fuzzier as to what their starting and ending states are.

The four kinds of labor

Here's how I generally name and think through these categories.

Factory Labor / Serfdom (high time, high task)

Some things are both highly time- and task-delimited. The example par excellence is the academic test. In a typical test, you have a specific amount of time, and a specific start and end state you want to leave the test in. Walk in -> take test -> hand in -> walk out. The preparation for the test is more like Salary Labor (low time/high task, measured against internal confidence that you've studied enough), but the test itself is Factory Labor.

Another example might be the manufacturing work done after a designer has gone through a few mockups and is ready to begin building a prototype. The goal: Create a prototype given blueprints and materials. You have a limited number of workshop hours you can spend in an average day to get them done. This task feels more task- than time-delimited to me, because you can run overtime, but it's pretty high on both axes.

I call this Factory Labor, or Serfdom if I'm feeling ornery. You have a specific number of widgets you have to crank out in your shift, so you work to accomplish that. Most people will end up cranking out about the same, average number of widgets, for the same, average amount of time.

Wage Labor / Meetings (High time; low task)

I work IT support as a part time job at my college. I have a set number of hours I have to be in for, and while some of the tasks contained within that time are Factory Work-esque, I'm generally being paid to be there in case I'm needed.

I call highly time-delimited, low task-delimited work like this Wage Labor for that reason. But much like Factory Work, don't let the name fool you: Not all work that falls in this category is in the service industry. Another name I considered was Meetings, because well-organized meetings are highly time-delimited, but often not highly task-delimited: There is an hour blocked for you to discuss whatever you bring to the table.

Salary Labor (Low time; high task)

How common is it that we have things we know we have to get done, but we don't know exactly how long they will take? If you ask me or my dad, pretty often, actually. Past a certain point, it actually becomes quite difficult to accurately estimate how long a thing will take on its own. I can estimate that each question on my real analysis homework will take me about an hour to solve, but adding up six uniform distributions makes me a lot more nervous about saying that the homework in total will take about six hours.

Now we're in the territory of Salary Labor. Salaried workers have a huge advantage in that their paychecks tend to be extremely regular; the tradeoff, of course, is that quite frequently the work takes longer than an ordinary 9-to-5 to get done. You have to stay overtime, and you don't necessarily know how long that will take.

Labor of Love (Low time; low task)

Say you're practicing guitar. You're pretty serious about it; you'd like to form a rock band maybe, someday, but for now you're just satisfied with becoming a better guitarist.

That's ... Actually a pretty vague category, though isn't it? Like, are we talking "technical death metal" better, or are we talking "blues throwback" better? I think most musicians would find it kind of a weird thing to try to put a box around in general, honestly, even if they could go all in on practicing specifics that add up to the goal, like finger picking or learning scales or the like.

In addition, unlike what Malcolm Gladwell likes to say, there isn't actually some magic "10,000 Hour" number you have to pass before you get certified as a Trve Kvlt Gvitarist. Really, you can't walk into these kinds of things with much of an idea at all about how long overall it'll take you. Best you can do is say "I'll practice for half an hour a day", but even then, there will be days where you play a lot more than that. (Gigs, for instance.)

This is what I call the Labor of Love quadrant. It's actually my favorite of the 4, and the one I'm most inclined to follow with my personal pursuits -- but it's almost never the best path to making money, due to the sheer amount of ambiguity around everything. How long will this take? Oh, you know. Will it at least be good? Might, might not. Aaaaaaaah! Just give your boss a fucking answer!


1 comments, sorted by Click to highlight new comments since: Today at 11:49 AM
New Comment

Cool 2x2. Are there any particular ways that doing a categorization of a task affects your work flow?

New to LessWrong?