Editorial policy

The project does not attribute

This record is published by a private owner acting as steward. The owner is not a specialist in Old Master prints, and neither is the software. Nothing here authenticates anything.

What the project does is narrower and, we think, more useful: it records what sources say, states how good each source is, and marks how much the evidence actually supports. Where an auction house has catalogued a sheet as by Rembrandt, this record says so and attributes the statement to the auction house — it does not adopt the statement as its own.

Every claim carries its evidence

A claim on this site has four things attached to it: the evidence supporting it, a confidence state, the date it was added, and the date it was last checked. If a claim has no evidence, it is recorded at unknown and says so plainly.

Confidence describes the quality of the evidence and nothing else. This is enforced mechanically rather than by good intentions: the build refuses to compile if a claim's confidence outranks the tier of its strongest source. A claim cannot be moved to confirmed by deciding it is true.

Uncertainty is published, not smoothed

Unknown, probable, disputed and rejected are all first-class states. A field that nobody has examined says nobody has examined it. A gap in the provenance is drawn at its true length. Where two institutional records disagree — and on this print they do — the disagreement is published rather than resolved in whichever direction suits the object.

Nothing is silently corrected

Superseded claims are deprecated, not deleted, and the reason is logged. The changelog is a credibility feature, not an embarrassment: a record with no corrections in it is either very new or not being read by anyone who knows better.

Price is not published

The project does not publish what the object cost, does not publish valuations, and does not frame the object in market terms. This is not squeamishness. A scholarly record that carries a figure invites every reader to weigh the scholarship against it, and the object becomes an argument about money instead of a research question. The rule is enforced by a build gate that scans the source and the data for currency figures and market language and refuses to deploy if it finds any.

Who can change the canonical record

At this stage, one person: the steward, acting on evidence submitted by others and credited to them. That is a weakness and it is stated as one. The intention is a small named group of external reviewers with the standing to overrule the owner on matters of cataloguing, conservation and provenance, and the version 1.0 gate list requires at least one external reviewer before the dossier is deposited with a DOI.

Contributions

Submitted corrections and evidence are reviewed before anything enters the canonical record. Nothing is edited into the record directly by a submitter. Accepted contributions are credited by name in the changelog and, at the contributor's option, against the claim itself.