Working document: last updated 29 August 2026 / kfe
This document draws on a technical discussion among DRL contributors in August 2026, which helped us clarify the wider infrastructure required to preserve endangered cryptoart and locate the work already underway within that larger picture.
A token is not an artwork. It is a permanent record of ownership that points to files stored somewhere else, usually on infrastructure the platform paid for. When a platform closes, those files can become unreachable while the token remains perfectly intact — and this is the normal case.
The scale of the cryptoart conservation problem is enormous. Platforms are closing or withdrawing services faster than endangered collections can be assessed and secured. Conserving this work requires multiple layers of effort: we must acquire the underlying data, preserve it in multiple independent locations, reconstruct any platform-specific behaviour needed to experience the art, and establish systems for continuing stewardship.
This document attempts to separate those layers, state the problems within each one, record DRL's solutions already in progress, and identify some outstanding questions.
<aside> 🧭
A note on scope. It's beyond any organisation's capacity to pin all cryptoart, and DRL isn't attempting it. What we propose to do is articulate the problem, set standards, offer resources, communicate what is happening, and perhaps map what is being conserved where. DRL is proposing action at two levels: encouraging individuals to secure their own holdings, and pursuing platform-level recovery.
</aside>
Conservation: The overarching ethical and practical framework for safeguarding artworks over time. In cryptoart, this may include acquiring endangered data, preservation, restoration, access, documentation, and continuing stewardship. Conservation also requires that decisions, interventions, uncertainties and reconstructed elements are recorded transparently.
Distributed conservation: As a guiding principle, DRL treats decentralisation as a conservation strategy. We aim to reduce reliance on any single platform, provider, device, institution or person by developing independent copies, shared documentation and transferable stewardship. The long-term goal is a system capable of surviving both technical failures and changes in personnel.
Archive: An organised body of media, metadata, records and contextual information gathered and maintained so that works can be identified, studied and potentially reconstructed in the future.
Preserve: Keep the actual data intact, retrievable and verifiable over time. Preservation requires maintained copies, not merely a record of where a file once existed.
CID: A Content Identifier: the cryptographic address or fingerprint of a particular piece of content on IPFS. A CID identifies the exact content, but does not itself contain or preserve the file.
Pin: Instruct an IPFS node or pinning service to retain and supply the data associated with a CID. A CID without an available copy of its underlying data cannot be repinned.
Restore: Re-establish the artwork as faithfully as possible, including the relationships among its token, media, metadata, configuration and intended behaviour. Any reconstructed or substituted elements should be documented.
Serve / provide access: Make preserved or reconstructed work available through a viewer, gallery or other interface. Data can be preserved offline without being publicly served; conversely, a functioning viewer can remain online while the files it depends upon are precarious.
Redundancy: Maintain multiple independently controlled copies so that the loss of one account, provider, node, device, institution or location does not destroy access to the work.
We can't preserve what we don't possess. As platforms continue to close or withdraw services, the most urgent task is acquiring the actual media, metadata and contextual records while they are still retrievable.