Early access to new essay: "Timeful Texts"

Thank you for those thoughtful observations, Phil. I think you're very right that different people will benefit from different prompts here. I'm interested both in designing the author-provided prompts to be somewhat more flexible (hope to write about that in the coming weeks) and also in crowdsourcing prompts. Your point about context also rings quite true, though I'm not yet sure how to operationalize it! One "dumb" perspective is that if reviewing the prompts can be made "cheap" enough, you don't have to do any clever targeting: stochasticity will do the work for you. But I'm sure there are interesting possibilities attainable in that space.

Andy Matuschak

Ha, what a coincidence! And yes yes let's explore implementation intentions with Orbit sometime ;)

Nicky Case

Hey Andy, Some thoughts comparing your concept of Timeful Texts to Richard Hamming's chapter on Creativity. 1) Different readers will have different interpretations the same work. It seems like an effective timeful system will allow readers to add their own custom reminders or pull from a common pool of community reminders. There is likely a status game in here for having the most widely referenced timeful summary of a book - look at how people flaunt their Goodreads pages! 2) Hamming calls these analogies "hooks" and talks about processes for creating more of them. Based on whatever is the top problem in your mind, timeful texts will result in different inspiration. E.g. if I am thinking about a hard engineering problem, and am sent a timeful text on my recent reading of philosophy, I will attempt to apply it to the engineering problem. There is probably a relationship between timeful texts and their relevance to certain problems, although I would be wary of becoming too prescriptive about this (ie only show engineering texts when working on engineering problems) -- most new innovations come from combining fields that seemingly have nothing in common! Can't wait to hear more. Thanks for the great work, as usual.

This is very helpful indeed, Nicky! You might be amused to hear that the examples of timeful texts in the piece originally focused on implementation intentions! But in the end, we felt that we just hadn't done enough prior experimentation with such prompts ourselves—the ideas are still too raw. I'll need a good context to explore those ideas more deeply (and perhaps that will be a collaboration with you!) I probably should have clarified in the post above: this piece has already been submitted to a print publication, so I can't make revisions except for correcting outright errors. But I'm grateful for your suggestions anyhow!

Andy Matuschak

Peter, I'm very grateful for this thoughtful engagement! I love your observation about Taleb. It reminds me of a point Hollis Robbins has made to me: that if something in a class makes you laugh out loud, or if it embarrasses you, you're very likely to remember it! So maybe we don't need all these fancy systems if we can simply construct more emotionally-laden prose. :) But I tend to believe that the effects are additive. And I believe that the ongoing social component is essential in Taleb's case: he's part of some low background hum of cultural gestalt. If you do find a podcast aggregator like you describe, please do let me know! Thanks, too, for that very kind conclusion. 🙇‍♂️

Andy Matuschak

Hi Andy, this is great! It's mostly stuff you & Michael have already written about in QC and TFTT, but it's good to see a compressed, standalone version of it. Also, the sine wave the images make through the page is *gorgeous*. Three suggestions/feedbacks – two low-level, one medium-level: 1) "When a user remembers or forgets the answer to a prompt, the system expands or contracts that prompt’s repetition interval along an exponential curve." Though I'm familiar with spaced repetition, this phrase still really tripped me up. Because 1) an exponential curve can't *contract* unless it's exponential decay, and 2) it's not natural to think of a *gap* exponentially growing, rather than a quantity. Might I recommend rephrasing it to be more concrete? Like, "Let's say it's been 5 days since you last saw a prompt. If you remembered it, it'll more than double the gap, and show it to you again in 14 days. If you forgot it, it'll less than half the gap, and show it to you again in 2 days. Thus, these expanding intervals allow you to maintain a collection of thousands of prompts while reviewing only a few dozen each day." 2) Speaking of writing more concrete, I think the essay could be more visceral if, before saying "The most powerful books reach beyond their pages", you have a short paragraph *listing* a few such books. Like: "The Analects of Confucius. The Feynman Lectures. The Elements of Style. The most powerful books reach beyond their pages—beyond those few hours in which they’re read—and indelibly transform how serious readers see the world." 3) Since you mentioned meditation, and since recently I've been reading more about Aristotelian virtue ethics, I wonder if it's worth mentioning a possible application of SR: using it to train "implementation intentions", i.e. habits, i.e. *character*. Like described in this post: https://www.lesswrong.com/posts/iocMeRcSKf3gAxsTh/self-programming-through-spaced-repetition Hope that feedback's helpful! Excited to see when it launches for real :)

Nicky Case

Nice! I wrote out some quick thoughts: https://docs.google.com/document/d/1f9cV7HSUq1UwvidQE6bCrBgNnuS7QivJUjNIv0M6At4/edit

Peter

I loved Maggie's illustrations too. Most of the ideas in here were familiar to me but they came together really beautifully in this piece. "Its goal is not simply to transmit reference material. The book aspires to convey a strong sense of how to think like X" is such a compelling vision to me.

Can Sar

These illustrations are 🔥🔥🔥