Source policy

The archive's usefulness rests on the quality of what it cites. This page records, plainly, which sources the archive draws from and which it refuses, so that any reader can audit how a given claim was arrived at.

Allowed sources for text claims

Text claims draw from: institutional archives (the Computer History Museum, the U.S. National Archives and Records Administration, the Library of Congress, the Smithsonian, the Internet Archive, museum and university collections); peer-reviewed historical literature (the IEEE Annals of the History of Computing, university press monographs on computing and printing history); primary sources where they are available (patents, manufacturer datasheets, contemporaneous trade press); and well-established encyclopedic references (Wikipedia, where its citations themselves resolve to one of the source categories above).

Allowed sources for images

Images draw exclusively from sources whose licensing posture is unambiguous: Wikimedia Commons (with the file's license string recorded verbatim), NARA, the Library of Congress "Free to Use and Reuse" sets, Smithsonian Open Access, the Internet Archive, and the Computer History Museum's public collections. Public-domain works (U.S. Federal Government works, works whose copyright has lapsed, works dedicated to the public domain) and compatible Creative Commons licenses (CC0, CC BY, CC BY-SA) are accepted; license-incompatible material is refused even when it would be visually preferable.

Forbidden sources

The archive does not source from stock-photo platforms (Alamy, Getty, Adobe Stock, iStock, Dreamstime, Shutterstock), content aggregators or social platforms (Pinterest, image-hosting aggregators without verifiable provenance), AI-generated imagery, fake retro recreations, or unverifiable personal blogs. The principle is not aesthetic: it is that the licensing chain on these sources is either non-existent or non-auditable, which makes them unsuitable for an archive that asks readers to trust its provenance claims.

License discipline

Every image carries an explicit license string recorded verbatim from the source institution's descriptor page. CC-BY-SA images are flagged as such; public-domain works are flagged as such; works released under a non-CC institutional rights statement (such as Flickr Commons's "No known copyright restrictions") are recorded with that exact phrasing rather than translated into a different license. The archive does not attempt to relicense or paraphrase license terms.

Per-image metadata requirements

Every committed image must carry: descriptive alt-text, exact pixel dimensions, a source attribution line, a license string, and where available a URL pointing to the source institution's descriptor page (not the upload URL). The repository's build pipeline enforces this with a content-integrity gate; an image that fails any of these checks fails the build, and the entry cannot ship until the metadata is repaired.