# Tree Beard

> A 2001 Central Saint Martins dissertation by J. Hudson-Powell proposing that web browsing be traced
> as a tree grown in three dimensions, recovered in 2026 from a Macromedia Director projector
> and republished with the original software running again.

Canonical page: https://jody.xyz/treebeard/
The essay as plain text: https://jody.xyz/treebeard/essay.txt

## What it argued

Browsers of 2001 recorded history as a flat list and offered a Back button, discarding the three
cues by which people actually re-find things: time, space and movement. The dissertation proposes a
companion program that traces a session as a tree the user turns and inspects. Each site becomes
a node. Branch length encodes time spent, damped by length = c x e^-kt so late branches stay in
proportion. Branch width encodes habit, so repeated routes thicken into a trunk. A site reached
again by another route is never duplicated; a translucent link is drawn instead. Leaves mark
favoured sites; fruit appears when a site has changed; growth can be pruned, taken as a cutting
and replanted, or seeded from a keyword and left to grow unattended.

The argument is developed through memory and recollection, the linear nature of language, growth
and emergence, cartography, and information-foraging theory.

## Provenance and editorial status

- Written and built c. 2001 in Macromedia Director 8 for Classic Mac OS. Not a document about an
  interactive piece: it IS the piece, read by clicking nodes rather than turning pages.
- The text was recovered in 2026 by decompressing the projector binary. The published text is
  edited for reading (spelling, punctuation, some long sentences split, and three
  claims narrowed or removed that the author no longer stands behind). The
  unaltered recovered text is the document of record and is held in the author's archive.
- Two quotations used without credit in 2001 are now attributed: Deleuze and Guattari,
  'A Thousand Plateaus'; Emile Durkheim, 'Les formes elementaires de la vie religieuse'.
- The original software was decompiled with ProjectorRays and ported to JavaScript. The engine,
  node coordinates, tooltips and page wiring are the author's own recovered Lingo.
- The work is NOT a priority claim. Comparable systems predate it, including Xerox PARC's
  hyperbolic browser (1996), Andrews' Harmony (1995), and PadPrints (1998), which already
  encoded dwell time and visit frequency. RSS shipped in 1999. The author says so on the page.
- ORIGIN: the idea did not come from the literature. The author was teaching himself Flash
  ActionScript and Director Lingo and working on interactivity and generative systems; the tree
  came out of that. He did not know the research field existed until a friend, Chris Lawrence,
  then studying computer science at UCL, recognised what he was describing and pointed him to
  it. The whole bibliography was found after that conversation.
- Of the 36 scripts recovered from the projector, 10 are the author's own, including the
  141-line 3D engine; the rest is Director's stock behaviour library.

## Reading list

Only four of the twenty-three references are named in the essay body (Kelly; Card; Pirolli;
Xerox). The rest were read rather than cited, which makes the bibliography a record of what was
on the desk in 2001 rather than a citation apparatus.

- Out of Control: The New Biology of Machines — Kevin Kelly, 1994 (Book). Wired's founding executive editor on systems that run themselves — swarms, ecologies, economies, machines that behave like organisms. https://kk.org/outofcontrol/contents.php
- Paul Virilio: From Modernism to Hypermodernism and Beyond — ed. John Armitage, 2000 (Book). A reader on the French theorist of speed, acceleration and the accident — the thinker who argued that every technology invents its own catastrophe. https://openlibrary.org/books/OL3965037M
- On Growth and Form — D'Arcy Wentworth Thompson, 1917 (Book). The classic of morphology: biological form as the product of physical and mathematical law, growth described by equation rather than by drawing. https://www.gutenberg.org/ebooks/55264
- Maps and Their Makers — G. R. Crone, 1953 (Book). A standard short history of cartography, from portolan charts to the survey. https://archive.org/details/mapstheirmakersi00cron
- Elements of Cartography — Arthur H. Robinson, 1953 (Book). The discipline's foundational textbook, in print since 1953 — the book that made the map-as-communication the governing idea of modern cartography. https://openlibrary.org/books/OL5687145M
- Maps and Politics — Jeremy Black, 1997 (Book). The map as an argument rather than a record: what gets centred, what gets named, what is left off. https://reaktionbooks.co.uk/work/maps-and-politics
- Visual Explanations: Images and Quantities, Evidence and Narrative — Edward R. Tufte, 1997 (Book). The third of Tufte’s books on information display, and the one about showing process, motion and cause — how a picture can carry an argument. https://www.edwardtufte.com/book/visual-explanations-images-and-quantities-evidence-and-narrative/
- History of Cartography — Leo Bagrow, revised by R. A. Skelton, 1964 (Book). The scholarly history, translated from Bagrow’s German and revised by the British Museum’s map curator. https://openlibrary.org/works/OL4590748W
- A Brief History of Time — Stephen Hawking, 1988 (Book). The popular cosmology that sat on every shelf in the 1990s. https://openlibrary.org/books/OL2402980M/A_brief_history_of_time
- The Holographic Universe — Michael Talbot, 1991 (Book). The bestseller arguing that reality and memory are holographic — that every fragment contains the whole, so a memory has no single location. https://archive.org/details/holographicunive0000mich
- The Hyperbolic Browser: A Focus+Context Technique for Visualizing Large Hierarchies — John Lamping & Ramana Rao, 1996 (Paper). A tree laid out on the hyperbolic plane: whatever you focus on is drawn large, and the rest of the structure shrinks toward the edge without ever disappearing. https://www.cs.kent.edu/~jmaletic/cs63903/papers/Lamping96.pdf
- Galaxy of News — Earl Rennison, 1994 (Paper). News stories rendered as a starfield you fly through, clustering into constellations by relationship as you approach. https://dl.acm.org/doi/10.1145/192426.192429
- Measuring the Web — Tim Bray, 1996 (Paper). Bray — later co-editor of the XML specification — measured the web’s actual shape: how large, how connected, how deep it went. https://www.areto.ethz.ch/CDstore/www5/www134/overview.htm
- Visualising Cyberspace: Information Visualisation in the Harmony Internet Browser — Keith Andrews, 1995 (Paper). The Hyper-G browser that put a 2D structure map and a 3D information landscape beside the document view, tightly coupled so that moving in one moved the other. https://ftp.isds.tugraz.at/pub/papers/ivis95.pdf
- Narcissus: Visualising Information — Hendley, Drew, Wood & Beale, 1995 (Paper). Documents drawn in three dimensions where position is not authored but emergent: objects attract and repel along their links until a structure settles out. https://ieeexplore.ieee.org/document/528691
- The WebBook and the Web Forager — Card, Robertson & York, 1996 (Paper). Web pages bound into 3D books and held in a virtual room, so a reader could keep dozens of documents within arm’s reach instead of one at a time. https://dl.acm.org/doi/10.1145/238386.238446
- The STARLIGHT Information Visualisation System — Risch, Rex, Dowson, Walters, May & Moon, 1997 (Paper). Large-collection visualisation built for intelligence analysis: documents as points in space, similarity rendered as distance. https://ieeexplore.ieee.org/document/626486/
- SemNet: Three-Dimensional Graphic Representations of Large Knowledge Bases — Fairchild, Poltrock & Furnas, 1988 (Paper). Knowledge bases drawn as 3D directed graphs, with fisheye views and clustering to keep them readable. https://www.taylorfrancis.com/chapters/edit/10.4324/9780203782040-5/semnet-three-dimensional-graphic-representations-large-knowledge-bases-kim-fairchild-steven-poltrock-george-furnas
- Navigating Large Networks with Hierarchies — Stephen Eick & Graham Wills, 1993 (Paper). How to draw a network too large to draw: impose a hierarchy on it and let the viewer descend through the levels. https://ieeexplore.ieee.org/document/398870/
- Visualizing Network Data — Becker, Eick & Wilks, 1995 (Paper). The canonical picture of network traffic: arcs thrown over a map, line weight standing for volume. https://dl.acm.org/doi/10.1109/2945.468391
- Cost-of-Knowledge Characteristic Function — Card, Pirolli & Mackinlay, 1994 (Paper). A way of measuring what a display costs its user: how much effort per item of information gained. https://dl.acm.org/doi/abs/10.1145/191666.191753
- WebView: A Graphical Aid for Revisiting Web Pages — Cockburn, Greenberg, McKenzie, Jasonsmith & Kaasten, 1999 (Paper). Thumbnails of the pages you have already seen, arranged inside the browser so you can get back to them. https://www.csse.canterbury.ac.nz/andrew.cockburn/papers/webViewOzCHI.pdf
- Revisitation Patterns in World Wide Web Navigation — Linda Tauscher & Saul Greenberg, 1997 (Paper). The study that logged what people actually did in a browser, rather than what they said they did. https://grouplab.cpsc.ucalgary.ca/grouplab/uploads/Publications/Publications/1997-RevisitationPatterns.CHI.pdf

## Web references, 2001

Eight of the nine addresses cited are dead, and each is linked on the page to an
Internet Archive capture from 2001 or thereabouts. GTrace is the only one still sitting
at its original address. The NicheWorks address as cited was never captured; the link
goes to the same page at a slightly different address.

- http://www.world.isg.de/World/2_Internet/Visual_Web/index.html — Innovative Software — Visual Web. A German software house selling a web visualisation tool.
- http://www.mappa.mundi.net — Mappa.Mundi Magazine. Carl Malamud's magazine about mapping the internet, and where Martin Dodge's Atlas of Cyberspaces lived. That atlas is probably where most of this came from.
- http://www.caida.org/tools/visualization/gtrace/ — GTrace — A Graphical Traceroute. CAIDA's graphical front end to traceroute, drawing the path packets take across the world. The only one of these still sitting at its original address.
- http://www.bak.spc.org/iod — I/O/D — The Web Stalker. The London group whose fourth release was The Web Stalker, a browser that refused to render pages and drew the link structure of a site instead. Software is mind control, get some.
- http://www.ibm.com/java/mapuccino — IBM alphaWorks — Mapuccino. A Java tool out of IBM's research incubator that crawled a site and drew its links as a map.
- http://www.cs.bell-labs.com/user/gwills/NICHEguide/niche.html — NicheWorks — Bell Labs. Graham Wills's guide to NicheWorks at Bell Labs, for drawing very large graphs. The address I cited was never captured. The same page on bell-labs.com, without the cs, was — and that is where this goes.
- http://www.neotrace.com — NeoTrace — NeoWorx. A consumer traceroute that drew the route to a host over a world map.
- http://www.tracemap.mids.org/test/tracemap/ — Matrix.Net tracemap. Traceroute maps offered as a web service, in beta. Paste in a route, get a picture back.
- http://www.inventix.com — Inventix Software — Internet Cartographer. Internet Cartographer built a category map of the sites you visited. Of everything on this page it is the closest to a product someone actually shipped on the idea I was proposing.

## Citation

J. Hudson-Powell, 'Tree Beard', Central Saint Martins College of Art and Design, London, 2001.
Recovered and republished at https://jody.xyz/treebeard/, August 2026.