Greatidude

ஆசிரிய கொள்கை / Editorial Policy

Editorial Policy

Effective 2026-04-23

1. What we are — and what we are not

Greatidude is a data-journalism project. We are not a fact-checking organisation, we do not issue binary true / false verdicts, and we do not claim editorial endorsement of any political party, alliance, ideology, or candidate. Our role is to convert public records into a standard set of comparable indicators and to publish the methodology that does the conversion.

2. Separation of data from opinion

Every indicator on the Platform belongs to one of two categories:

Data Sourced from government portals; displayed with a link to the primary source and a capture timestamp. E.g. attendance percentage, criminal cases on record, fund utilisation, bills introduced.
Opinion Derived from disclosed methodology; labelled as such. E.g. Gap Score, "Overrated" / "Underrated" classification, "Dark Horse" badge. Based on public records; opinion is constitutionally protected but is not a statement of fact.

Promise statuses that assert a verifiable outcome (KEPT, BROKEN) are shown publicly only when supported by a linked primary source. Absent such support, the public display is NEEDS_REVIEW regardless of any internal draft classification.

3. Methodology transparency

Our scoring formulas are published at /how-it-works. Their version history is at /methodology/changelog. Every politician profile, and every social card derived from one, includes a methodology link in the same surface area as the score.

4. Equal application

The same formulas, thresholds, and data sources are applied to every politician regardless of party, coalition, seniority, or current media prominence. When a new politician is added to the Platform, backfill is scored under the current methodology version and tagged accordingly.

5. Pre-publication checklist

Before any automated content (video, newsletter, social post) whose gap band falls into OVERRATED or CRITICAL_OVERRATED is distributed, the following are verified and logged in content_approvals:

  • At least one primary-source URL is cited in the asset itself and in its public description.
  • The methodology URL is present in the same surface.
  • No KEPT or BROKEN promise label without an evidence URL appears in the asset.
  • No comparative phrasing attributes motive (e.g. "corrupt", "lazy"). Comparative language must be measurable ("lower attendance than N% of peers in the same session").
  • Model Code of Conduct status is checked against ELECTION_SILENCE_MODE.
  • Two reviewers (one of whom must be the Editor or Legal reviewer) have independently signed off with APPROVED in content_approvals.
  • A right-of-reply pointer to the relevant politician profile is included in the asset's description.

6. Right of reply

Any politician or their authorised representative may submit a verified response on the politician profile page. Verified responses appear inline with the score card, are not edited, and carry the submitter's name, role, and date.

7. Corrections

Every public field change on the Platform is recorded in /corrections: the field, the before and after values, the date, and the reason. Corrections arising from grievances are cross-referenced to their grievance reference code.

8. Model Code of Conduct (MCC)

When the Election Commission of India notifies an MCC-silence window covering Tamil Nadu (or any state in scope of the Platform), automated content generation and distribution are paused. The platform continues to serve existing data read-only. A banner discloses the pause.

9. Conflict of interest

Contributors and advisors are required to disclose any current or recent political, governmental, or commercial relationship that could materially influence coverage. Disclosures are published at /transparency. No member of the team has held a position in any political party or government body at the time of writing.

10. Advisory review

Methodology changes of MAJOR scope are reviewed by an external advisory panel — a political scientist, a Tamil-medium journalist, and an election-data researcher — whose public sign-off accompanies the release. Advisors' bios are published at /transparency as the panel is seated.