Writing experiments and the banana escape valve

post by Dmitry Vaintrob (dmitry-vaintrob) · 2025-01-23T13:11:24.215Z · LW · GW · 1 comments

Contents

1 comment

[Note: This is not alignment-related, but rather a spacefiller personal blog post.]

I've been trying to write a public post every day [LW · GW] of January. So far I’ve been enjoying it. I don’t think this approach works for everyone: in particular, I’ve also been hanging on by a thread to the schedule and to the ability to sleep. I publicly committed to write 3 posts a week, but my “secret” personal goal to write a post every day for the month. Not only in order to have more output, but because in the past intense daily deadlines have worked pretty well in the past; also there’s a kind of scientific idea of “you’ll get more data on the success of a new behavior if you go the whole hog”.

In order to have a clear “success criterion”, my rules are that 

• By day n of January (for $n\le 31$), I want to have authored or coauthored n posts or shortforms in January (including the first announcement post).
  • Here the “end of the day” is defined as Pacific “Random standard time” – so January 1st ends at 6AM January 2nd.
• During any 7-day period, I’ve written at least 3 posts (this is unlikely to lapse before the previous condition, but is there to avoid cheating by something like “writing 31 short posts in one day”).
• There is no restriction on quality, length or content (other than a vague “have an internal sense that you are not hurting the commons by posting overly personal or unproductive schlock on a public forum. Graded leniently, ok?). In my personal precommitment on fatebook, I gave myself a 5% chance of success. It’s a bit higher now but still under 50%.

I’ve actually done a couple of experiments of this type before, and have enjoyed them. The latest was three years ago, when I gave myself a “daily writing minimum” of 12 pages in a small notebook, for a month. It was a conference-heavy month and while at first the pages had discussions of literature and math and productivity:

The last page looked like this:

For projects like this one, I feel like it's very useful to have consistent practices to phone it in: graceful failure modes that allows you to retain steam. In the context of this sprint, I actually rough-drafted a couple of escape valves (including this post) in advance, in case of an emergency backup.

Other practices I've used this sprint:

• converting old drafts or conversations into low-effort posts
• writing what are essentially math lectures
• taking the first idea or coherent set of ideas introduced in a post and splitting it off as a "first part". (I particularly recommend this one.)

I think that for some contexts and people, giving yourself goals that are easy to goodheart creates bad incentives. But in other contexts having a “banana-themed escape valve” as above is exactly what the doctor ordered. You get a ping of shame whenever you use the valve, but it also lets you maintain a sense of continuity and intactness of your project. I think in this way the hack is similar to religion: sometimes keeping the sabbath means pretending your keys are a brooch or tie-clip, but the continuity this engenders makes it more meaningful when you actually use it for its intended purpose of untangling yourself from earthly affairs and communing with the divine.

If I were to guess, there are two general types of motivational structures that people have that let them get stuff done. One is like money: you care about efficiency and total output, and the other is like religion, where what matters is consistency, ritual, “wholeness”. These combine differently for different people, and I’m definitely more on the hippie dippie ritual side of the divide. Similarly for tasks, I think this kind of incentive structure works better for tasks where you inherently have an identity-linked sense of “wanting to get it right” rather than mundane “unrewarding” tasks like doing taxes (egosyntonic and egodystonic are probably the correct psychobabble terms). 

I’m planning to do a more in-depth postmortem of the writing project once the month ends. Meanwhile, I’m thinking of new (less time-intensive) monthly projects to run in the near future. Stay tuned for a post about this later. Thanks to any readers for reading!

Banana.

1 comments

Comments sorted by top scores.