Still going analog
Last week I wrote about using my phone much less and trying to give myself more and longer blocks of no-screen time. I'm still at it:
- I'm still turning off my phone during work hours. My messages and phone calls all come to my computer, so I'm not missing anything urgent. I'm just not using computers during coffee breaks and the like.
- Guiding long-running LLM activities, including in the mornings and evenings, is part of programming now. This limits how long my analog blocks can be. Even bounding my screen time this way (going analog before and after my first and last work sessions) is a big change.1
- Tracking daily to-do lists on paper, where I can revise them and integrate them with notes, is remarkably effective. Of course, any number of digital tools let you integrate tasks with notes and edit them freely (Obsidian is great!). Even so, I'm finding it surprisingly useful and pleasant to plan my days by hand. I at least feel myself prioritizing and planning more effectively in "analog headspace."2
- Reading books has felt much different, and better. So much of my experience of a book happens when I'm not reading it: fiction, especially, not only spurs explicit contemplation but also passively colors so many day-to-day experiences. I suspect that checking my phone soon after a reading session interferes with that.
- This leads to a pessimistic conjecture about the decline of book-reading. The standard explanations, I take it, are that (i) people have a considered preference for other things (short-form video, video games, or whatever) or (ii) they akratically choose something with an immediate hit even if they think books are better for them. A compatible but different idea is that books are actually worse when combined with many patterns of smartphone and other digital usage, because the away-from-book aspect of reading gets worse.3
- I've felt sharper both at work and in (e.g.) flashcarding. But my flashcarding data do not support the hypothesis that I'm actually performing better (or its opposite). It's hard to measure the effect of this kind of systemic lifestyle change, largely because so many things change at once (e.g., the time of day when I'm drilling).
- I am sleeping better, but the sample size is tiny, and cause and effect are hard to sort out here.
- The whole experience is more Proustian than I expected. For example: it's been hot here, and simply walking a quarter-mile through the heat without a phone is triggering vivid memories of 2006 Las Vegas and 2002 State College.
For all that's written about "LLM psychosis" and compulsive or addictive AI usage, I haven't seen much about the psychological effects of using chatbot-adjacent AI tools on the job for many hours a day (but non-compulsively). I am (i) only one person, (ii) not a doctor, and (iii) doing unusual and AI-intensive work, but I'm still finding that even modest increases in analog time feel very good.
A lot of my colleagues (and people in other professions) really dislike the idea of checking in on work outside "9 to 5" hours. It doesn't bother me, and even if it did, I suspect it's simply a new requirement of high performance in what I do.↩
The obvious question here is, whether or not there are benefits to doing this non-digitally, how high the costs are. My current answer is: not much. It's easy to digitize what needs digitizing later, and I suspect that I've long overestimated the benefits of these digital records. But that's another post.↩
I think the standard view of watching sports is roughly analogous: watching a game or even part of a game is very different when you check your phone between plays. The big differences are that (i) sporting events are intrinsically ephemeral in a way that many books aren't and (ii) your phone is more likely to be showing you something about the game you're watching than the book you're reading. Everyone recognizes these effects, but I don't see as much analysis of them as I'd like. (Ben Thompson and John Gruber are welcome exceptions here.)↩