Encyclopædia Mundi Encyclopædia Mundi Encyclopædia Mundi Encyclopædia Mundi

MFA Thesis, 2021.

Encyclopædia Mundi is a studio art installation built around a fake 1990s encyclopedia program, styled like something you’d find on an old Windows 98 machine. You search a topic, get a real opening line pulled from Wikipedia, then a chunk of text generated by GPT-2 that sounds grammatically fine but slowly goes off the rails — repeating itself, contradicting itself, getting weirdly obsessed with random details — before snapping back to a real Wikipedia sentence at the end. The whole thing comes wrapped in a fake backstory: a poster claiming the software was found at a yard sale, in-app notes from “developers” at a fictional failing tech company. The first reaction is usually “wait, is this real?” before the cracks show.

The point isn’t to trick people so much as to poke at how we decide what counts as trustworthy knowledge: encyclopedias rarely credit individual authors, so their authority comes from the format and presentation itself, not any named expert. Dressing up nonsense in that same authoritative packaging asks what happens to “truth” once AI starts generating text that’s fluent but hollow. It ties into an ongoing interest in tech obsolescence — this time software going obscure rather than hardware decaying. Influences include Monty Python’s Dirty Hungarian Phrasebook sketch, Lauren McCarthy’s 24H HOST, and Windows93 (windows93.net).

Technical implementation

The piece has two main components. The client — the audience-facing, Windows-98-like application — was built with Electron (Node.js), using jdan/98.css (github.com). Windows 98 was installed in a virtual machine so pixels could be counted, leading to changes in the original CSS to reach pixel accuracy for a win32 application circa the late 90s. Because the entire application is a rendered viewport, basic features like dragging the window by its title bar had to be reimplemented over IPC — down to details like double-clicking the title bar icon quitting the app, a behavior dating back to Windows 3.1.

When the audience searches a topic, the client sends the query to the server over the network. The backend was built with Flask (Python), which returns the first two sentences of the closest-matching Wikipedia article. The first sentence is passed to the GPT-2 tokenizer and a pretrained model on the server, instructed to generate a set amount of text (an earlier version delegated this computation to an instance on GCP). The client streams the real first sentence out a character at a time, masking the time GPT-2 needs to generate the fictitious continuation — a trick to hold the audience’s attention while maintaining the illusion of an obscure, outdated program.