Editorial method

Every published record carries a direct answer before any acronym, mission number, or archival vocabulary; at least one source assertion with its source class and a link to the source; explicit date precision; and a stated relationship to the other records it touches. A record missing any of these is not held for review — it is simply not publishable, and the import that carries it is rejected.

Objects carry a further field that ordinary articles do not: what the object shows. Stating the claim an image supports is what keeps it from quietly becoming an argument it cannot carry.

Where sources disagree, the disagreement is published rather than resolved silently. Where a date is known only to a year or a decade, the precision is stated next to the date, and date-derived views such as Today will not treat it as an anniversary.

Some records are deliberately held rather than published: unresolved source conflicts, sensitive history, tragedy, a living person’s likeness without recorded clearance, an unsourced quotation, uncertain rights, unlabelled representation, or incomplete provenance. Each hold names its specific reason, so it can be resolved rather than argued about.

Every image carries a truth class at the point you meet it — archive, restored, colorized, diagram, reconstruction, or present day — as a word, not only a colour. Source character is preserved: borders, frame numbers, legends, and catalog marks stay when they carry evidence, and no artificial aging is ever added.

Synthetic imagery is never used as historical evidence. Where a generated illustration fills a genuine visual gap it is labelled on the frame itself, and it never fabricates a real person’s likeness, a quotation, a document, or a date.

Articles are published under the Space Now byline with J. S. Nova as the accountable editor. Corrections are published as corrections: what changed, why, and when. Records are superseded by new revisions rather than edited in place, and the previous revision is retained.

It does not generate answers with a language model, summarise records automatically, or offer a chat interface. It does not require an account, an email address, an app, or cookie consent before you can read, search, or inspect a source. It does not publish a reverse-chronological blog, and it does not promise a daily or weekly cadence it has not committed to.