Source before conclusion.
Architecture before assertion.
Record, Not Rumor is a custom-built public evidence system. It preserves source provenance, separates evidence from allegation and inference, tests public claims against their strongest available record, exposes evidentiary limits, and makes material corrections visible.
What Record, Not Rumor is.
Record, Not Rumor is an independently built records-audit and claim-verification application. Its founding archive concerns the Clancy case, but the system is designed around a reusable evidence standard rather than a true-crime format.
Claims are decomposed into testable propositions and compared with authenticated or otherwise properly classified source material.
Sources are identified by what they are, where they came from, what context surrounds them, and what they can and cannot establish.
Published findings expose the source path, reasoning boundary, evidentiary limits, and correction path rather than asking readers to trust the creator.
Subjects may change. Source hierarchy, chronology controls, classification rules, correction standards, and public-access commitments do not.
What it is not.
Record, Not Rumor is not a crowdsourcing platform. Future public submissions may provide leads or source material, but popularity, volume, voting, follower count, community consensus, or repeated submission never determines evidentiary weight or a published conclusion.
It is not a true-crime publication. A criminal case may be the subject of a bounded record; sensationalism is not the editorial model.
It is not a no-code or template publication. The current production application is custom-built and maintained as source-controlled software with automated verification and deployment controls.
It is not a paid truth layer. Published evidence audits, source methodology, record matrices, and material corrections are not placed behind a reader paywall.
Where did you get this?
That question comes before: “What does it say?”
- Identify the submitted artifact. State whether it is a court record, transcript, exhibit, news article, screenshot, social post, video, audio recording, database entry, correspondence, or another source class.
- Identify the origin. Provide the original URL, publisher or issuing body, author or speaker when known, publication or filing date, and enough information to retrieve the source independently.
- Trace screenshots and excerpts backward. A screenshot of text is treated as a lead until the underlying article, filing, transcript, recording, or other origin can be located and its context checked.
- Evaluate the source itself. Determine whether it is primary, official, authenticated, contemporaneous, expert evidence, attributed reporting, argument, allegation, commentary, or an unresolved artifact.
- Follow its citations. When an article or secondary source cites a filing, study, transcript, exhibit, interview, dataset, or other record, the review continues to that underlying source whenever reasonably available.
- Preserve context. Read or watch enough surrounding material to prevent a cropped quotation, attorney question, headline, edit, or screenshot from silently becoming the meaning of the complete source.
- Cross-check independently. Test the proposition against chronology, other witnesses, records, physical or digital evidence, explicit source limitations, and material evidence that could disprove the working conclusion.
- Classify and expose the limit. Publish what is supported, what is contradicted, what is inference, what remains unknown, and what evidence would be required to change the finding.
What is true? What is false? What is still unknown?
The reviewed record establishes the proposition within the stated source boundary.
Reliable source material directly contradicts the proposition as circulated.
Material components are supported while other components are false, unsupported, or unresolved. Each component is identified separately.
The source supports a narrower statement than the public claim, or a real fact has been combined with an inference that the source cannot establish.
No authenticated or otherwise properly supported source reviewed within the stated boundary establishes the proposition. This is not a claim of omniscience.
The available evidence cannot responsibly determine the proposition. Uncertainty is preserved rather than converted into a verdict.
A better source can change the finding.
If authenticated or materially stronger evidence supports, contradicts, or narrows a published conclusion, Record, Not Rumor will revise the classification and explain what changed and why.
The creator is not the evidentiary source. The creator designs and applies the review standard. A finding must therefore be capable of surviving a challenge from a better source—and capable of changing when it does not.
Custom software. Reviewable controls.
The purpose of this disclosure is verifiability—not exposure of protected implementation details.
Transparency is not an operator manual.
Public governance describes controls and guarantees without publishing credentials, environment variables, private repository locations, internal project identifiers, security tokens, unpublished source locations, or other operational details that would create unnecessary attack surface.
- The public intake form remains disabled until private delivery, spam controls, retention boundaries, and abuse handling are verified.
- Public evidence access does not require exposing protected source code.
- Security controls are proportional to the current public/static attack surface and are revised when the architecture changes.
Source truth is not a premium feature.
Record, Not Rumor does not charge readers for access to its published evidentiary audits, record matrices, source methodology, or material corrections. Evidence quality is not determined by purchasing power, membership, or subscriber status.
Future professional services associated with the creator are separate from this public evidence project and do not alter the evidentiary standard or public-access rule.