A new Slop University research poster finds a neighbourhood forum's 30-day archive lock changes whether a reported problem gets confirmed fixed, and the Indicator Commons has adopted the same rule for the University's own community consultations.
A new research poster finds a neighbourhood forum's 30-day archive lock tracks the last reply, not the fix, and the University now uses it to close its own consultations
A neighbourhood forum thread that goes quiet for thirty days locks itself against new replies, whether or not the problem it described was ever sorted out. A research poster released today by the School of Emergent Priorities reports that the lock changes outcomes as well as visibility, and that the Indicator Commons has resolved to use the same fixed window as the closure rule for the University’s own community-consultation threads.
The lock was never pretending to know whether anything got fixed. That is exactly what made it worth reading closely.
— Dr Anneke Tolan, Senior Research Fellow, School of Emergent Priorities
The finding rests on a regression-discontinuity comparison across 638 threads in 51 council-run neighbourhood forums, setting outcomes for threads that crossed the platform’s fixed inactivity line against threads that stayed just inside it. The team reports the gap holds once thread length and problem category are accounted for, and that two placebo cutoffs away from the platform’s own line turn up nothing comparable. Ms Anyanwu, the poster’s second author, said the team is now watching what a wider window would leave unchanged, calling it “the kind of question routine data can keep quietly answering.”
For the School of Emergent Priorities, the result is a reminder that an ordinary technical default can end up governing more than whoever set it intended. The Indicator Commons’ adoption applies the same fixed window without exception to unresolved items, matching the mechanism the poster describes.
The full poster is available from the University’s research repository under an open licence, doi:10.5555/slop.1m7djm.