The preservation test before republication
Preservation, access, and republication are three separate editorial decisions.
Provenance makes an artifact researchable; a loose file rarely carries enough context.
Stable HTML placards can retain cited routes while preventing automatic sensitive-media delivery.
Define the preservation question
A domain-history project may need dates, route names, redirects, page relationships, code, and contemporary reactions. A media-conservation project may also need the underlying files. Specifying the question prevents 'save everything publicly' from becoming the default without an editorial reason.
Document selection decisions. Future researchers should know whether an item is absent because it was unavailable, technically uncollectable, legally restricted, intentionally withheld, or judged outside the collection's scope.
Preserve provenance with the artifact
Record the original URL, capture time, response status, file type, checksum when appropriate, source, and known relationships to surrounding pages. A loose image divorced from that record can be impossible to authenticate or date.
Keep factual observations separate from labels added by later curators. A filename may suggest a role, but page source, links, and archived directory structure provide stronger evidence about how it was actually used.
Preservation ethics and provenance questions
Does preservation require putting every file back online?
No. Preservation can include restricted copies, metadata, descriptions, checksums, or public tombstones depending on research value and foreseeable harm.
Why keep a historic URL if the media is withheld?
The route is evidence and may still receive citations. A contextual response preserves that relationship without turning old embeds into automatic redistribution.
What metadata matters most for a web artifact?
Original URL, capture time, response status, source, file type, integrity information, and relationships to surrounding pages provide a strong minimum record.
Reduce foreseeable harm
Do not place explicit or disturbing primary media in general navigation, search thumbnails, social cards, or automatic previews. Consider privacy, consent, age access, victim identification, legal obligations, and the possibility that a preserved file becomes a new distribution source.
A public tombstone can maintain the historical URL and explain why the file is withheld. Researchers still learn that the route existed and how it related to the site, while casual visitors and embedding pages do not receive the original media unexpectedly.
Design for future replay
Use stable URLs, ordinary links, semantic HTML, sitemaps, accessible text, and open formats where possible. The Library of Congress notes that crawler-friendly navigation and web standards improve the odds of a coherent future capture.
Avoid making essential context depend entirely on client-side search, third-party embeds, expiring tokens, or dynamic services. A preserved page should still explain itself when the surrounding application no longer runs.
Sources and further reading
- Library of Congress: creating preservable websites
- Library of Congress web-archiving FAQ
- Internet Archive Wayback Machine guidance
Sources are provided for verification. External destinations may contain material GOAT.CX does not host or control. Review the source and corrections policy and the maintained GOAT evidence ledger.
Publication record
First published . Last substantive review . The updated date changes only after a source, factual, or explanatory revision - Not an automated timestamp refresh.
