compost.cc is permacomputing[1] for common digital archives — a media art project by Zachary Furste and Diego Gómez-Venegas. The project restores discarded computing devices into servers that collectively preserve endangered archives. The network distributes files using custom software based on the BitTorrent protocol. Have a look at a live network terminal.

collection #1

The first collection is documentation from Chile's Project Cybersyn (1971–1973) — a cybernetic system for managing the national economy, built under Salvador Allende and destroyed by the 1973 military coup. The archive has been scattered across institutions and private collections for fifty years.

This collection includes only documentation that was clearly produced — and therefore its rights are owned — by the state of Chile through one of its agencies — either extinct or still active. By international standards and Chilean law, these documents are public information.

collection #2

The second collection is the archive of Screen Walks, a collaborative project by Fotomuseum Winterthur and The Photographers’ Gallery investigating the changing role of the photographic image in its networked and digital forms. Artists, curators and researchers were invited to perform live-streamed explorations of the digital spaces where their core practice takes place. The first Screen Walk with Penelope Umbrico took place on 8 April 2020, and the last one with Ting-Chun Liu and Leon-Etienne Khür on 26 March 2025.

Across five rungs, the compost.cc collection holds original and compressed video recordings of the 86 Screen Walks and additional performances, a guided tour and a workshop, along with audio versions, transcripts, and preparatory and archival materials.

The programme was launched by Jon Uriarte, then digital programmes curator at The Photographers’ Gallery, and Marco de Mutiis, digital curator at Fotomuseum Winterthur.

September 2026

Screen Walks collection now seeding

We are delighted to announce the second collection, and the first external deposit, on compost.cc: the archive of Screen Walks.

A pioneering programme that began during the COVID-19 pandemic, Screen Walks was a series of 86 live-streamed walks through the digital spaces where artists, curators and researchers work, recorded between 2020 and 2025. The programme was launched and curated by Jon Uriarte of The Photographers’ Gallery and Marco de Mutiis of Fotomuseum Winterthur.

The collection deposited on compost.cc is available across five rungs, from compressed video and audio to original recordings, transcripts and archival materials.

Watch the announcement video here and then install a node to join in on the fun.

upcoming

21–27 September 2026

Unplugged?

Participation in Bucharest, sponsored by Marginal and V2_ Lab for the Unstable Media.

Workshop: Bring Your Own E-Waste, Leave with an Archival Server The Legacy House, Bucharest, Romania

date: 13:00 26 September 2026

15 October 2026, 15:30 (Mexico City time)

Association of Internet Researchers (AoIR) Conference

Lecture-performance by compost.cc in Mexico City.

27 October 2026

University of Basel Digital Humanities Lab

compost.cc lecture and workshop.

past

17 September 2026

Open Tabs Club - The Web We Want - Error 417, The Photographer's Gallery

Presentation of compost.cc

24 June 2026

Critical Computing Online Gathering

Online presentation of compost.cc.

7–10 May 2026

[FUTURE] TERRITORIES Transmedia Exhibition — Das LOT × NODE Media Lab

compost.cc installation at das LOT, Vienna.

Bring Your Own Device workshop: Saturday, 9 May, 15:00-18:00.

the problem

Many important cultural archives lack visibility. Some are carefully preserved by archival staff without the knowledge or resources to host digital copies. Others disappear. Institutions lose funding, servers go offline, hard drives fail, political regimes change. When a digital archive depends on a single organization or a single server, it is only as durable as that organization’s budget or that server’s uptime.

compost.cc invites people to share in the work of preservation — distributing archives across many machines in many places, maintained by the people who care about them. Our interest is not in decentralization for its own sake, but in collective responsibility: more hands, more resilience, more of a sense for what it takes to keep shared knowledge alive.

how it works

Each node in the network runs a single small program that does everything: it checks a signed list of what the network is currently preserving, downloads the relevant files, and shares them with every other node using BitTorrent. The list is cryptographically signed, meaning that each node can verify it hasn’t been tampered with before accepting changes.

  compost.cc                          node (rescued device)

				  ┌──────────────┐                    ┌──────────────┐
				  │  registry    │── fetch+verify ──▸ │  compost     │
				  │  .json       │                    │  binary      │
				  │  + signature │                    │              │
				  └──────────────┘                    │  embedded    │
				  ┌──────────────┐                    │  BitTorrent  │
				  │  tracker     │◂── announce ────── │  engine      │
				  └──────────────┘                    │              │
				  ┌──────────────┐                    └─────┬────────┘
				  │  web seed    │◂── fallback ───────────┤
				  └──────────────┘                          │
				  ┌──────────────┐                    ┌─────▾────────┐
				  │  directory   │◂── status ──────── │  BitTorrent  │
				  │              │──▸ peer hints ───▸ │  swarm       │
				  └──────────────┘                    └──────────────┘

Once files are downloaded, nodes share them directly with each other, peer to peer. A tracker helps nodes find each other. The more nodes that join, the more resilient and available the archive becomes.

trust

Two separate cryptographic keys govern the network. An archive’s curator signs a manifest attesting that these are authentic documents with verified provenance. The network operator signs the registry that tells nodes what to seed and where to find it.

This separation means archival authority and infrastructure authority are independent. A new curator can take over a collection without needing access to the infrastructure, and a new operator can manage the network without needing access to the archive.

the devices

The network runs on rescued hardware like laptops pulled from e-waste, tablets in drawers, and phones with cracked screens, restored with lightweight Linux and turned into archival servers. We see turning discarded technology into infrastructure for preserving discarded knowledge as compost.cc's central gesture.

The node software is designed for these machines: a single ~20 MB binary using about 40 MB of RAM, x86/ARM/MIPS Linux, no technical configuration beyond the initial install. Node operators choose how much to seed based on their available disk space — even a minimal contribution from an old phone strengthens the archive’s resilience. The network also includes servers provided by eclips.is and the Digital Methods Initiative.

permacomputing the commons

compost.cc operates under the philosophy of the commons. These archives are not owned. Instead, they are collectively maintained. The project also draws from permacomputing — a framework for digital practices that takes seriously the material and ecological costs of computation, prioritizing care, repair, and the extended use of existing resources over replacement and upgrade cycles. compost.cc insists that hardware is worth repairing and that infrastructure should be legible.

concessions

in a spirit of transparency, we acknowledge the following concessions to the ideals laid out above:

--we have used LLMs, including those provided by major companies like Anthropic and OpenAI, while writing the code used in the compost.cc software and infrastructure (and also for the German, Dutch, and Romanian translations of this 'about' page);

--we have used ISPs, major cloud service providers, and video conferencing software that are out of step with permacomputing principles while developing compost.cc;

--we have used closed-source software for some tasks in producing and processing images related to the project;

--much of the development and creative work related to the project was carried out on machines which were purchased new within the last 6 years;

--we have booked roundtrip flights (including one transatlantic) for the purpose of presentations and workshops connected to the project.

we welcome suggestions and discussion about these choices and other tensions that we have not yet thought to list.

participate

Run a node. If you have an old machine — even a low-powered one — it can become part of the archive. Install the node software and send us an email for an invite code.

Bring a device to a workshop. compost.cc hosts hands-on sessions at cultural institutions where participants bring old devices and leave with a functioning archival server. No prior technical knowledge required.

Donate hardware. Old phones, tablets, or laptops — especially those that can run Linux. Get in touch.

Host a workshop or exhibition. If you run an institution and this interests you, reach out.

Browse the code. Everything is open source at codeberg.org.

Above all, reach out: hola@compost.cc.

Demo installation of five recycled devices running compost.cc nodes: a 2011 MacBook Air with a broken screen, a 2012 MacBook Pro, a 2018 Huawei MediaPad T5 with a broken screen, a 2014 ASUS MeMO Pad, and a Google Pixel 7 with a broken screen, alongside a Wi-Fi router and a power meter.