Week 3: Code as Exhibit

Titaÿna Kauffmann, C²DH — Luxembourg Centre for Contemporary and Digital History, University of Luxembourg (Exhibition Organizing Committee)

Pierre Depaz, Postdoctoral Researcher, Universität Basel (Exhibition Scientific Committee)

On January 29, 2026, Software Heritage and Inria will open a new exhibition at UNESCO Headquarters in Paris. The "Source Code Exhibition" presents fifteen code examples spanning seventy-five years of computing history. Each is interpreted by a different contributor—historians, artists, developers, activists. The exhibition takes an unusual approach: it treats source code as the central exhibit, not merely as contextual material for hardware or interfaces.

We are part of the organizing team, and we'd like to bring the project to this community: not as a finished argument, but as an invitation to critique, extend, and challenge the choices we've made.

Software Heritage: Preservation as Foundation

This exhibition grows from the work of Software Heritage, a nonprofit launched in 2016 with an ambitious mission: collect, preserve, and share all publicly available source code. Their archive now holds over 18 billion unique source files from more than 300 million projects—a Library of Alexandria for software.

But Software Heritage's vision goes beyond storage. As they put it: "Software in source code form is produced by humans and is understandable by them; as such it is an important part of our heritage that we should not lose." Code carries our technical and scientific knowledge. It mediates access to all our digital information—lose the software, and we lose the capacity to interpret what we've created. Software Heritage recognizes that preserving code is essential for preserving cultural heritage itself.

The exhibition marks Software Heritage's tenth anniversary. It experiments with what becomes possible when preservation meets interpretation. An archive makes code available; an exhibition asks what it means.

The Three Axes

The exhibition organizes its fifteen panels around three overlapping axes:

Source code as historical testimony — how recovered code offers windows into technical and social transformations, from the first programming textbook to ELIZA to XMODEM.

Source code as mirror of society — how code reflects and shapes the cultures that produce it, from a Saturday afternoon git commit to RSA encryption on a protest T-shirt.

Source code as cultural artifact — how code functions as an expressive, aesthetic form, from the thirteen-character fork bomb to programming in classical Chinese.

These aren't rigid categories; most panels resonate across multiple axes.

The Exhibition Online

The physical exhibition space at UNESCO limited us to fifteen panels. We received fifty-one proposals in response to our call for contributions. Selecting among them was difficult. We sought diversity—in time periods, programming languages, geographic origins, and contributors' relationships to code. The final selection reflects those priorities, but also inevitable compromises.

Our hope is that the online version will offer more flexibility. All panel texts, code examples, and contributor bios are available under a Creative Commons license. We aim at revisiting some of the initial proposals and welcome new contributions in the future. We encourage you to spend time with the materials before diving into discussion.

Visit: www.sourcecode-exhibition.softwareheritage.org

Starting the Conversation

Over the coming days, we'll post in-depth code critiques of exhibition panels for discussion, alongside a practical introduction to using Software Heritage as a research tool. But we wanted to open this week by inviting reactions to the exhibition as a whole—its premise, its choices, its gaps.

For instance: how do we best communicate the cultural significance of source code to a broader audience? Many visitors at UNESCO won't have written a line of code. What works, and what gets lost in translation? To paraphrase Benjamin, what is the exhibition value of source code?

Are the three axes we chose—historical testimony, mirror of society, cultural artifact—the right categories? Do they help or hinder interpretation? Are there any other categories that might seem relevant?

And more broadly: what can code learn from existing heritage practices, and what new heritage practices might code require?

We've also posted a companion thread on Software Heritage as a resource for code studies—how to navigate the archive, what's possible with their tools and APIs, and how researchers in this community might use it in their own work.

We look forward to hearing your thoughts—and to exploring the exhibition together.