Sourcing and verification.

Every photograph in Suede DNA originates from the guitargeek-archives item on the Internet Archive, a community-maintained compendium of guitarist rig diagrams built up over years, much of it originally surfaced through guitargeek.com and related rig-documentation communities. Suede DNA does not photograph rigs, does not commission photography, and does not re-host the underlying images on its own infrastructure. Rig images are served directly from archive.org/download/guitargeek-archives/.

What Suede DNA adds is a presentation, normalization, and search layer over that upstream collection, not a re-publication of new material.

The canonical dataset is data/rigs.json, built by a script (scripts/build-manifest.ts) that parses upstream filenames against a known grammar, artist_guitar_rig_year.ext, and produces a flat list of rig records: an id, an artist slug, a year, a source URL, and a file format. A companion script (scripts/validate-manifest.ts) checks the manifest's schema and integrity before every build; a broken or malformed manifest fails the build rather than shipping.

Two small hand-maintained files sit on top of the automated parse: data/overrides.json, which corrects display names where the upstream filename grammar is ambiguous or abbreviated (turning ac_dc into AC/DC, for instance), and data/excluded.json, a short list of items intentionally left out, mostly upstream files that returned an error status when fetched and could not be resolved to a valid image.

It is worth being precise here, because “verified” can mean very different things. The manifest pipeline verifies structure: that a rig record has a valid artist, a plausible year, a working source URL, and a consistent id. It does not verify content. Suede DNA does not interpret what is inside a rig photo. There is no gear-list reconstruction, no signal-chain transcription, no cross-referencing against interviews or magazine features. Each rig is presented as the source document it is: one archival photo, one year, one artist, and a link back to the upstream file. The compilation is the value here, not an editorial interpretation layered on top of it.

That also means Suede DNA does not independently fact-check the accuracy of the upstream archive itself. If a photo in guitargeek-archives was mislabeled with the wrong year or the wrong artist before it reached this site, that error can carry through. The overrides.json file exists specifically to catch and correct cases like that as they are found. It is a living correction log, not a guarantee that every record is error-free.

The archive documents 409 rigs across 390 artists. That is a meaningful sample of guitar history, not an exhaustive one. Many artists have no entry at all; many who do have only one or two documented years out of a much longer career. Absence from the archive says nothing about an artist's significance. It only reflects what the upstream guitargeek-archives collection happened to document and what has been normalized into the manifest so far.

If you find a misattributed rig, a wrong year, or a broken image, the fastest path to a fix is @aisuede on X or info@suedeai.org. Corrections are applied through data/overrides.json. The underlying image is never edited or re-hosted, only the metadata around it.