DRMacIver's Notebook

Projects that you can’t run out of

Projects that you can’t run out of

Published
2022-01-21

I was working on my new consulting website yesterday. I can’t say it’s the most exciting thing in the world, but one of the interesting things about working on it was that it gave me a project of a sort which I don’t think I have that often, and that felt quite good.

I don’t have a good term for such projects, but let me see if I can articulate what I’m pointing at. They have the following characteristics:

  1. They can usefully absorb pretty much as much work as you’re willing to put into them - they may plateau, but the work never becomes pointless or harmful.
  2. They don’t have to - you can stop pretty quickly and walk away and you’ll have done something good.
  3. What you’ve done stays done.
  4. They feel neutral to positive to work on.

For example writing a blog isn’t this, because it fails (1) - after a certain level of effort, a post starts getting worse rather than better, and if you write too much you just overwhelm the audience (people who want to keep up are already behind on reading my notebook, and I’m only writing daily! That’s no more than an hour of work a day).

Projects that take a very long time to pay off also aren’t this. e.g. a PhD, or even a paper, very much isn’t this because if you walk away from it before it’s successful, you’ve failed.

Cleaning and general maintenance aren’t this, partly because you have an obligation to keep on top of them - you can definitely fail to clean - and partly because they fail (3) and just undo themselves.

A lot of paid work fails at (4) but more interestingly paid work also often fails at (1). Paid work fairly rarely is something you can sink all your time into, because it starts to feel really bad to do that eventually - unless you’re billing by the hour, you’re being exploited. It goes against norms of work-life balance. Your coworkers resent you making them look bad. That sort of thing. Sometimes your employer wants you to put unbounded amounts of effort into your paid work, but that’s a different (and usually worse) problem which rarely results in positive feelings.

Projects like this are really satisfying to me. I think it’s partly because they don’t trigger the difficulty finishing things but there’s also something very satisfying about knowing that you can just give something your all and will be rewarded for it. They give you a reliable way of turning hard work into positive feelings, and are usually a ready source of flow, and that’s pretty great.

One of the interesting things about such projects is that, contra my recent line of thinking, they’re not very good as team projects, because the problem is that if you’re reliant on other people for working on something then you can’t put unbounded amounts of work into it - you can put work until you get to something where you’re blocked on other people for. I think this is part of why personal software projects are often much more satisfying (or at least differently satisfying) than shared projects.

Some other examples of work like this for me are Hypothesis (when I was still working on it) and, for a more trivial recent example, working on my Wordle solver. They’re interesting problems that can be endlessly tinkered with to improve upon, but that I very quickly got to a point where it’s good that they exist and would still be good even if I walked away from them now.

Another interesting, darker, class of examples is that I think everything that gets used to drift has this character, but in a less nourishing way. I think the key difference is something like short-term vs long-term positive feelings. If I spend time playing a game, that produces good short-term feelings. Depending on how much time it may also cause good feelings in retrospect. e.g. I genuinely enjoy Slay the Spire and think it’s worth the time I spend on it, but if I sank 80 hours a week into playing it, I would not feel good about that looking back.

Perhaps drift is, partly, looking for something of this sort and, lacking a good source of it, turning to the least bad ones available to us?