The trust manual

Editorial Standards

Sharp elbows. Straight facts. The joke never gets to counterfeit the evidence.

Receipts over rumors

Material factual claims should link to primary documents, public records, full transcripts, official data, or direct statements whenever those sources are available. Secondary reporting is identified as such. A viral screenshot is a lead, not proof.

Reporting, analysis, and satire are different jobs

We label the format and do not smuggle opinion into a sentence dressed like a fact. Satirical language may be biting; quotations, figures, timelines, and source descriptions may not be invented.

Plain English, full context

We explain what happened, what the evidence establishes, what remains uncertain, and why the distinction matters. We do not confuse complexity with sophistication or certainty with confidence theater.

Corrections

Substantive errors are corrected in the article with a dated note describing the change. Small spelling and formatting repairs may be fixed silently when they do not change meaning. We do not erase a bad call and pretend the internet had a hallucination.

Pundit and outlet scoring

Scores measure the reviewed sample, not a human soul. Reviews consider factual accuracy, source transparency, correction behavior, context, certainty calibration, and the gap between headline or on-air confidence and available evidence. Reviews state their source window and can be revisited.

Conflicts and independence

Relevant financial, political, or personal conflicts should be disclosed. Sponsors, if any, do not preview or edit coverage. Paid material will be conspicuously labeled. Access is not editorial control.

Evidence has limits

When documents are incomplete, disputed, unavailable, or inconclusive, we say so. “We do not know yet” is a complete sentence and often the only honest headline.