the lab
essay

A reader's stance

On reading as an active, aggressive thing — and why the tools we have mostly assume otherwise.

October 12, 2025 ↳ margin tools for thought

There is a stance one takes when reading well. It is not the stance most software assumes you are taking. Most software assumes you are skimming for a quote to paste into a Slack thread. This is sometimes accurate, but planning for it as the default is a kind of design malpractice.

The stance is aggressive, in the way a careful sculptor is aggressive with stone. You arrive at a paragraph believing the author is mostly wrong and try to find what’s right anyway. You re-read the previous page when you discover, on this page, that you’d misunderstood. You write in the margin not because you’ll review the note later but because writing is how the reading completes itself.

A small list of things this stance requires of an interface.
  • It must let you write while reading, not after.
  • It must not move things around when you write.
  • It must let you write next to anything — a word, a line, a figure, a whole section, the gap between two paragraphs.
  • It must remember without asking. No save button.
  • It must let you find the note again three weeks later by remembering almost nothing about it.

My current claim is that this stance can be designed for, but not designed into; you can build a room that invites a posture, but you can’t make the reader sit up straight. This is part of what Margin is trying to figure out.

Adjacent: Reading on paper vs. screen, and the more practical thread in Notes vs. marks.