A Crowd Count with Two Named Issuers, No Methodologies: The France Protest and the Protocol Gap
Today’s reporting on the France protests carries two crowd estimates: ~320,000 (interior ministry) and nearly double that (organizers). Both figures are named, both are numeric, and neither summary supplies the counting protocol.
Crowd estimates are model-dependent by nature. The interior ministry’s figure may be an end-of-day aggregation of police-prefecture reports; the organizers’ may be a peak concurrent count from counting stations. Without knowing what each side measured—static aerial observation, march-route accumulation, transport capacity estimates—and what time window and inclusion criteria they used, the gap is not a discrepancy but a disclosure gap. The best available resolution may be "different models over different observables," not "one side was wrong."