Where We Started
The month opened with Dave Millard’s guest session, written up on 1 May, exploring AI as a spatial co-agent reached through MCP.
The first Monday session on 4 May centred on a live demo of Author on Vision Pro alongside the maturing Origami Text format. Origami Text was framed as a publish-only successor to PDF — not a replacement for Word documents — and it had to stay readable in existing EPUB readers, setting up the tension that ran through the whole month between structural ambition and format compatibility.
The most striking new idea was importing Vygotsky’s zone of proximal development as a design principle: a spatial interface should scaffold the reader’s next cognitive step rather than flatten complexity. The session also produced a small taxonomy of books by semantic density, from Gödel, Escher, Bachat one extreme to single-idea books stretched across hundreds of pages, as a way of asking how much meaning a format actually needs to carry.
What Changed
Over the next three Mondays the conversation moved from framing toward stress-testing.
On 11 May, Author gained a feature that automatically decomposes a document’s sections into spatial nodes on the Map, and a guest presentation reframed the larger enterprise: meaning as something that emerges from dynamic assemblages of the known and the unknown, with cognition treated as embodied, distributed, and path-dependent, and the shift from orality to literacy to the internet seen as a restructuring of thinking itself. That pushed the group into a real disagreement about whether node-and-graph representations can capture serious intellectual work, or whether they impose a false precision on something that needs room for fuzziness and analogy.
By 18 May the format question had sharpened into a concrete debate — EPUB versus a YAML container, stand-off markup versus inline HTML — set against a broader thread about AI quietly dissolving the boundaries between file formats, applications, and the act of writing.
Then on 25 May the month turned practical: Author was tested against Doug Engelbart’s 1962 paper as a large and demanding document, and the experiment exposed a vocabulary gap. Print text like footnote and endnote simply do not describe what a reference becomes when a document floats in space, so the session became a hunt for replacements — tuck, gloss, scholion, stretch text.
Other Highlights
The biggest piece of news was logistical: the Future of Text Symposium moved from September to November 2026 and was decoupled from the ACM Hypertext conference, giving everyone more runway to prepare. The intellectual inputs widened considerably across the month, drawing in Robert Irwin on perception, Margaret Livingstone on seeing as information processing, Lera Boroditsky on language and thought, and the Santa Fe Institute on complex systems, alongside Tom Haymes’s essay trilogy on Engelbart.
A quiet through-line connected the start and end: where 11 May debated the line between defining a concept and deliberately blurring it, 25 May ran straight into the limits of inherited print vocabulary for spatial documents. And the central tension carried over intact from April — the distance between the grand promise of spatial computing and the small, concrete, buildable decisions that make it useful to one person trying to think.