How the record is kept.
Two things together: what canopticon is trying to do, and how it checks what it puts on a page. A claim from this site is trustworthy because both are visible, and both can be checked.
The standard
Every documented fact rests on a real source. Analysis is signed and kept separate from fact. Confidence is shown, not hidden — an uncertain claim says so. Nothing is stated that can't be traced back to a document. The record is permanent and correctable: mistakes are logged, not erased.
What's in scope isn't a topic list — it's a question: is more said about this than has ever actually been checked? Wherever that's true, it applies, whether the subject is a politician's record, whether a therapy actually works, or whether a species is really recovering.
Checked by machine, not vouched for.
A claim on this site is trusted because of what happened to it before it was published, not because of who wrote it. Every documented fact runs a chain of checks:
A source document is stored first, exactly as received. Facts are then read out of it, each one carrying the exact quote it came from.
Every quote on every page is a literal match against the stored document — checked by machine, byte for byte. Whitespace differences are tolerated; word substitutions are not. A quote that fails this check never reaches a page.
Before a signed analysis publishes, a different model than the one that drafted it reads each claim and answers one question: does the quote actually support this? The judge is always a different model family from whatever wrote the page — self-checking is not allowed.
A claim that fails either check is removed before publishing — never silently kept. Where it's useful to show, a removed claim stays visible on its file with a plain note: failed verification — removed.
Every receipt on every page opens to the verbatim text, the source document, and when it was read. Claims on this site are deep-linkable to their evidence.
Disputed stays disputed
A claim someone made ("X said Y") is never silently shown as an established fact. Disputed material is stored as disputed and shown as disputed. A signed analysis is required to show the facts that complicate its own case — it can't leave out what doesn't fit.
The stance
Forensic, not journalistic. The goal isn't to report what happened but to map what is — the structure beneath the event, the pattern beneath the story, the architecture beneath the policy. Individual facts matter because they reveal structure. Structure is the finding.
What's actually in the record
Every file grows from a typed record of facts read out of source documents:
- People and companies — the persistent subjects a file is opened on.
- Events — dated occurrences with verbatim quotes.
- Documented facts — sourced, established, undisputed statements.
- Disputed claims — what someone claims, with who said it and their stance, kept distinct from documented fact.
- Sources — the documents themselves; every fact's receipt points back to one.
- Official stories — widely-held accounts flagged when three or more separate documents push back on them.
The links between all of these are typed too — who's connected to what, when, and how. The record is queryable; every page is a rendering of it.
One file, two views
Every subject — a person, a company, an event, a widely-held story — has a single file with two views: on the record (documented facts with receipts, disputes shown as disputes, a timeline, connections) and, once one exists, a signed analysis — an argument over the same facts, with an intent badge, a confidence level, and the mandatory box showing what complicates it. The analysis isn't a separate document — it's the same file with a thesis pulling the facts into order.
Formats of a signed analysis
Once an analysis is signed, it's rendered into a few formats: a short thread, a video script, a shareable card with the single strongest fact, and a chapter outline. None of these are new writing — they're selections from the signed piece. The claims live on the analysis; the formats just re-arrange them.
The lenses
A small fixed set — not a framework canon. Applied selectively, not mechanically.
- Fact vs. framing
- Separate the thing said from the backdrop it's said against.
- Root-striking
- Trace a surface fact to the structural cause underneath.
- Steelman
- State the strongest version of the opposing case before answering it.
- Cui bono
- Ask who benefits from this being the question.
- Base-rate
- Check the claim against how often such things are actually true.
- Dispute or question the logic
- Argue the facts are wrong, or that the reasoning doesn't follow even if the facts are right.
The discipline
A person owns every verdict. The system surfaces reasons for and against — it never decides. Patterns are hypotheses until verified and signed. Evidence that cuts against a claim is shown, not buried. The analysis is worth reading precisely because it can be wrong, and says so when it might be.