ReadingRate · Answers · 2026
How do I build a second brain for reading?
The hard part isn't the system, it's maintaining it — most reading note systems collapse because capture and filing cost more effort than they return. A second brain for reading works when saving a passage takes seconds, tagging and linking happen automatically, and old notes resurface without you going looking. ReadingRate is built on that shape: speak or photograph a quote, get themes applied automatically, and see connections to what you read years ago.
There's an obvious analogy for readers: runners have Strava, lifters have training logs, and reading has apps that count finished books. Counting books is a strange measure — it treats a novel read in a day and one that took three months as the same event.
What a reader actually wants from a system is to keep and retrieve what the reading gave them. That's a different thing from a shelf of covers.
Disclosure: we make ReadingRate. This page says what the feature can't do as well as what it can.
What the system needs to do
Capture in seconds
If saving a passage interrupts reading, it stops happening by week three. Speaking a quote aloud or photographing the page keeps you in the book.
File itself
Manual tagging is the step every system dies on. Saved quotes get themes applied automatically, so retrieval works without you having maintained a taxonomy.
Link across books
Passages are compared by meaning across everything you've saved, so a line from a novel surfaces alongside one from a poem or a science book when they share an idea.
Resurface on its own
Notes you never revisit may as well not exist. Saved quotes come back on a spaced-repetition schedule so you meet them again before they've faded.
Watch instead
Why book counts and page counts both mislead, and what a reading archive needs to do to still be useful years later.
Why counting books is the wrong metric
Finishing a short, easy novel and finishing a long, difficult one are not the same achievement, but a books-read counter records them identically. Page counts aren't much better, since a page might carry a hundred words or three hundred.
ReadingRate measures a day's reading as words multiplied by the book's difficulty, giving a rough cognitive load rather than a book count. Being precise about it: the word figure is derived from the pages you logged and the book's own word count and length, so it's an informed estimate rather than a direct measurement. It's a far better proxy for effort than counting covers.
Retrieval is the whole point
The test of a reading system is whether you can answer a question like "where was the passage about relationship anarchy in The Dispossessed?" months later. If the answer is to flip through every underline, the system failed.
Automatic theme tagging is what makes that work, because it means retrieval doesn't depend on you having filed things correctly at capture time — the moment when you're least willing to do admin.
Not being locked in
A second brain you can't get out of isn't one. Quotes, notes, and saved conversations export to Notion and Markdown, and your library exports as CSV compatible with the common reading trackers.
This matters more for reading notes than for most data, because their value is cumulative and measured in decades.
Honest limits
- Automatic themes are a reasonable first pass, not a curator's judgement. They make things findable; they don't tell you what matters.
- Connections are drawn from passages you've saved, so a sparse library produces sparse connections.
- Cognitive load is words times difficulty, but the words are estimated from your logged pages rather than measured directly, and some books lack a real word count.
- The underline-aware photo capture is currently iOS only; the web transcribes a whole page.
- No system supplies the judgement about what's worth keeping. That's the part that can't be automated, and it's the part that makes the archive yours.
Start the archive
Capture quotes in seconds, get them tagged and linked automatically, and meet them again later.
Related questions
- How do I build a second brain for reading?
- Make capture take seconds, let tagging and linking happen automatically, and make old notes resurface on their own. Systems that require manual filing at capture time are the ones that get abandoned.
- Why do reading note systems fail?
- Because filing costs more than it returns. Capture competes with reading, and manual tagging is admin nobody keeps up. Automating those two steps is what makes a system survive.
- Is counting books read a good measure of reading?
- No. It treats a book read in a day and one that took three months as equivalent. Words read weighted by difficulty is a much better proxy for effort.
- Can I export my reading notes?
- Yes. Quotes, notes, and saved AI conversations export to Notion and Markdown, and your library exports as CSV.