Blind Replication Sprint

A Blind Replication Sprint, also called a Convergence Hackathon, is the hackathon archetype in which multiple independent teams work in isolation on a shared dataset using methodologically distinct approaches by design, with convergence of results as the success criterion rather than ranking against results. The archetype is the only one of the ten primary hackathon formats in which the team is not the unit of competition; the unit is the protocol itself, and the integrity of the result is established through cross-team agreement on methods that should not have produced spurious convergence. The Event Horizon Telescope's 2017 EHT Image Hackathon at the Black Hole Initiative and the seven-week 2018 imaging sprint that produced the M87* image are the canonical examples. See format-taxonomy and event-horizon-telescope.

The integrity mechanic that distinguishes the archetype — convergence as proof when no ground truth exists on the day of judgment — is the substance of integrity-through-convergence, the Manifesto principle that grounds the format. The four conditions that make the protocol work (genuine team isolation, methodological diversity by design, pre-registered parameter sweeps, and the convergence event as a designed dramatic moment) are operational rather than theoretical, and CASP's biennial protein-prediction protocol, the parallel ATLAS and CMS Higgs detection of July 2012, and LIGO's blind injection methodology all operate the same logic in different empirical domains. The format is not portable to weekend hackathons — seven weeks of team isolation requires institutional commitment — but the logic generalizes to any hackathon whose outputs the world cannot independently verify on the day of judging. See blind-analysis, methodological-diversity, embargo, and convergence-event.