One big roadblock for environments hoping to improve on the book with fancy interactive elements: they all require reading on a screen! And reading on a screen is almost universally terrible!
I've been getting increasingly worried about this—seems like a serious threat! (con't)
Conversation
For instance, expert readers generally read non-linearly, but object permanence issues really inhibit that on screens. (notes.andymatuschak.org/Maintaining_mu, notes.andymatuschak.org/Continuous-scr)
In many cases (eg. on e-readers) *performance* issues inhibit expert reading! Wild! (notes.andymatuschak.org/Poor_performan)
10
7
81
Of course there are on-screen features which enhance expert reading… but on balance, I'm pretty reliably a worse reader when reading on screens.
Not thrilled about needing to solve those problems in addition to all the tools-for-thought ones! What do you find promising here?
14
1
35
What if navigation wasn’t through arbitrary chunks (pages) but aligned with the structure of the content (sections > paragraphs > sentences)?
Too much effort goes into transforming trees into sequences (writing) and back (reading). What if we just kept the trees? (fig. and lit.)
3
1
2
Replying to
"API design" is harder than it looks; requires careful, explicit thought. notes.andymatuschak.org/Evergreen_note
Prose—even outline-oriented prose—doesn't naturally lend itself to that kind of composition.
Quote Tweet
Replying to
Fully agree. Can’t quite make the connection to my earlier tweet (assuming it’s a response).
Have you read about Robert E. Horn’s Information Mapping? A somewhat extreme approach that only applies in specific contexts, but insights on improving scanning and reference seem useful.


