Themes and Read & Write?
We are mostly focusing on the space of knowledge and what that might mean.
On comprehension and XR
The Greek episteme means literally overstand — to stand above and survey. The Old English understandan means to stand among.
In XR we can build environments where you both simultaneously. You can zoom out to the overview — the bird’s eye, the overstand — and zoom in to be surrounded by a cluster of connected nodes — the understand. The XR headsets are perhaps the first reading environments that lets a person move fluidly between those two epistemic postures.
Meeting Notes
We came round to a notion that ‘reading’ and ‘writing’ might not be the right terms for interacting in XR. Suggestions based on dialog with Claude (read/write in XR):
- Locognition — fits English phonology perfectly, meaning is guessable, has gravitas
- Toponoesis / Toponoetic — beautiful, slightly harder but rewarding
- Chorognosis — if you want to keep the Greek soul of the original
- Knowplace — if you want something anyone can say immediately without explanation
Topics
- Interaction with the knowledge in the room. (Headset is X-Ray for what is ‘really’ there). What to show, how to interact…
Transcript
AI Analysis
The 23 March 2026 session of the Future Text Lab gathered Frode Hegland, Tom Haymes, Brandel Zachernuk, and Peter Dimitrious for a wide-ranging discussion centred on the nature of knowledge interaction in XR — specifically what it means to navigate, manipulate, and “enter” information spatially, as distinct from reading or writing in any conventional sense. The meeting wove together a live demo of Frode’s Author spatial mapping tool on visionOS, Tom’s ongoing project to build a personal knowledge navigator from his photographed bookshelves and Kindle library, and substantial theoretical exchanges touching on flow theory, LLM output quantity, the design philosophy of Doug Engelbart, the materiality of paper, the emerging poverty of vocabulary around knowledge-in-XR, and the role of music and photography as alternative channels for entering knowledge.
AI: Main Topic
The primary focus was the question of what a new modality of knowledge interaction in XR actually is — and what to call it. Frode demonstrated updates to Author on Vision Pro, showing spatial concept-mapping features including category-based global selection, depth distribution, focus filtering, and a refined prompt system that extracts concepts, people, locations, and quotations from documents. Around this demo, the group developed a shared position: what Author enables is not reading, not writing, and not simply viewing a 3D concept map. Brandel articulated the clearest formulation — it is “reading like a map rather than reading like a book” — a topological and topographical appreciation of textual spatiality. Frode arrived, mid-conversation, at the formulation: “It is actually the room. Here is the knowledge. The headset just gives me access.” This reframing — from tool to environment, from interface to knowledge-space — became a quietly pivotal moment in the session.
AI: Highlights
Frode explicitly pointed to the neurological basis of long-term memory as an underpinning for the spatial approach, connecting the design of Author to findings from the book Behave by Robert Sapolsky. He highlighted the principle: to understand knowledge, you have to explore it — and that this maps directly onto how long-term memory is formed in the brain.
Tom introduced a Karl Schroeder quote in the chat — “A codex teaches the reader to think in places. An ebook teaches the reader to think in streams.” — and posed the question directly: “What is XR reading? Bits?”
Brandel shared his demo of Yung Chang Heavy Industries kinetic typography in XR (https://zachernuk.neocities.org/2026/yhchang/), noting that the sparseness of the environment — text on white, nothing else — was itself a meaningful compositional choice. He also shared a link to a post by Anisha Moonka (https://x.com/AnishA_Moonka/status/2035798600148758698) on the cognitive differences between reading on phones and reading on paper, which Peter flagged as new and important territory for him personally.
Peter noted in the chat: “raw capability vs. intuitive / pleasurable” — distilling the design tension that ran throughout.
Frode noted that the community is “truly scratching at the surface of developing a terminology and a language for working in a knowledge space” and proposed a possible future topic: “what is interacting with knowledge in XR? Just that. Beyond reading and writing.”
AI: Insights
Brandel’s observation that design is not merely polish but “the centuries-long project of rounding off the edges through decades of practice” represents a substantive reframing of the Engelbart debate. Where Engelbart prioritised capability, Frode prioritises feel and fluency — but Brandel located the deepest problem neither in capability nor polish alone, but in the absence of a cultivated, intergenerationally transmitted practice. This absence is precisely what makes XR knowledge tools so hard to evaluate: there is no century of use to have done the sanding.
The group converged on a diagnosis that current LLMs produce output at a rate and volume that structurally prevents flow states for most users. Brandel argued that the moment LLM output becomes surprising, the user is necessarily ejected from flow in order to “properly apprehend what’s really going on” — meaning that either people are not truly paying attention, or they are not actually getting much information. This is not a complaint about quality but about the mismatch between generation speed and human cognitive bandwidth, and suggests that the right design intervention is not better LLMs but a metering and pacing layer.
Tom’s framing of the Knowledge Navigator as being about the “spaces between ideas” — not the nodes but the relationships not yet articulated — introduced a complementary vector to Frode’s explicit ontology of concepts, people, and locations. The two approaches imply different epistemological assumptions: Frode’s is structuralist (named entities and their typed relations), Tom’s is phenomenological (the felt gap between known things).
Peter’s articulation of the music-vs-lecture split in his own cognition — that his brain simply switches modalities when music plays and stops processing semantic content — pointed to a genuine design question for Frode’s audio accompaniment practice. Rather than treating this as a problem, Frode reframed it: intentional misunderstanding and partial absorption may be appropriate for a medium designed to provoke rather than inform, and this led him to propose a new format — a brief musical summary as intro, the conversation itself, and a longer song as outro.
Tom’s invocation of Muybridge as a historical analogue for XR‘s undiscovered expressive possibilities was generative: photography revealed motion it could not have set out to capture. The implication is that the novel affordances of XR knowledge spaces may only become visible through sustained inhabitation rather than deliberate design.
Brandel’s analysis of the Apple business school case studies — physical printouts with sidebars, implicit and explicit hyperlinks, supplementary geopolitical context — gave a concrete, non-speculative instance of the “felt need for a spatial workspace.” He reframed the question entirely: not “what is the VR thing?” but “what is the ideal workspace, and what does it mean to have the right things ready to hand?” This is a design question with immediate, practical implications that does not require a headset to ask.
Frode’s concept of “visual metadata” — the practice of explicitly labelling the importance and nature of information, particularly in the absence of machine-readable publication dates and provenance in PDFs — surfaced as a small but significant infrastructural proposal: that better annotation practices at the point of creation are a prerequisite for meaningful personalisation or spatial organisation later.
The exchange about cave paintings — paintings shaped to fit the rock’s contours, animated by a flickering torch — proposed an archaeology of immersive knowledge environments that predates digital media by tens of thousands of years. The implication is that the desire to make knowledge spatial, embodied, and slightly animated is not a novelty of XR but a recurrent human impulse, and that the present moment is a new instantiation of a very old project.
Tom’s closing observation that three-dimensional media remains “virgin territory” — that so far the only form he has encountered in XR is the map — raises a genuine open question: what would a non-map, non-flat representation of knowledge in three-dimensional space look like? The group did not resolve this, and the question remained productively open.
AI: Resources Mentioned
Websites and links
Future Text Lab session page: https://futuretextlab.info/2026/03/21/march-23rd-26/ — shared by Frode Hegland.
Author spatial map on visionOS: https://www.augmentedtext.info/author-map-visionos — Frode Hegland’s tool, demonstrated live.
Brandel Zachernuk’s Yung Chang Heavy Industries kinetic typography XR demo: https://zachernuk.neocities.org/2026/yhchang/ — shared by Brandel Zachernuk.
Anisha Moonka on reading cognition (phone vs. paper): https://x.com/AnishA_Moonka/status/2035798600148758698 — shared by Brandel Zachernuk; discussed further by Peter Dimitrious.
Imagine AI image generation tool: www.imagine.art — used and shared by Frode Hegland.
Books
Behave by Robert Sapolsky — mentioned by Frode Hegland as having been recommended by community member Andrea; cited as the origin of his thinking on the neurological basis of long-term memory and its connection to spatial knowledge exploration.
Understanding Media by Marshall McLuhan — mentioned by Brandel Zachernuk, who noted having read it entirely in headset.
Almost Perfect (history of WordPerfect) — mentioned by Brandel Zachernuk in the context of early debates about what a word processor is for.
Ministry for the Future by Kim Stanley Robinson — mentioned by Brandel Zachernuk as a current read; briefly discussed by Tom Haymes.
People
Doug Engelbart — inventor and augmentation pioneer; his H-LAM-T taxonomy and “downhill skiing” analogy discussed at length. The story of Ted Nelson coining the pronunciation “Hamlet” for H-LAM-T was recounted by Frode Hegland.
Ted Nelson — hypertext pioneer; discussed in relation to coining “intertwingled” and his naming of H-LAM-T. No URL given.
Ken Perlin — mentioned by Frode Hegland as having said he does not work in XR; Brandel Zachernuk intends to try to arrange for him to speak to the group during a New York visit on the 19th–20th.
Nonny de la Peña — VR journalist and filmmaker; mentioned by Brandel Zachernuk as an example of polished VR work.
Jeremy Bailenson — Stanford VR researcher; mentioned by Brandel Zachernuk as producing brilliant but insufficiently polished work.
Karl Schroeder — quoted in the chat by Tom Haymes: “A codex teaches the reader to think in places. An ebook teaches the reader to think in streams.”
Anisha Moonka — researcher or commentator on reading and cognition; link shared by Brandel Zachernuk, noted by Peter Dimitrious.
Eadweard Muybridge — mentioned by Tom Haymes in the context of photography revealing motion as an analogy for XR discovering its own expressive affordances.
Paul Smart — identified by Frode Hegland as the person Brandel was trying to recall (“David Chapman, the fellow who was doing the stuff at Southampton“).
Andrea — community member (surname not given); credited by Frode Hegland for recommending Behave.
Technologies and tools
Author (visionOS) — Frode Hegland’s spatial document tool. Claude — used by Frode Hegland for concept extraction, prompt iteration, lyric writing, and terminology generation; used by Tom Haymes for organising bibliography data. Gemini — used by Tom Haymes for initial bibliography processing and personal planning. Notebook LM — mentioned by Tom Haymes for its podcast remix function. Vision Pro / visionOS — Apple headset platform used for Author demo. Imagine AI — image generation tool used by Frode Hegland. Zoom, Final Cut, YouTube, Internet Archive, Google Docs, Illustrator, InDesign, AutoCAD — referenced in various contexts.
Song
This song is also part of the end of the video record, above. Lyrics