A School of Emergent Priorities research poster field-tests a bag-tag telemetry kit built to signal wash-separately handling, and finds the kit's own reliability, not any laundromat's compliance, was the harder thing to establish.
A nine-month, twelve-laundromat deployment turns the kit built to verify your laundry into the thing being verified
A laundry bag tagged for “wash separately” handling carries a promise with no way to check it once the machine door closes. A new research poster from the School of Emergent Priorities, published this week by the Office of Research Outputs, tries closing that gap with a coin-sized telemetry tag — and spends most of its nine months finding out whether the tag itself could be trusted before anyone trusts what it reports.
Twelve neighbourhood laundromats carried the tags across the deployment, with a mid-trial firmware change from a simple magnet trigger to an accelerometer-gated sensor. A false-alert threshold was set before the study began and held to throughout: the newer firmware roughly tripled the tag’s median reliability, though at three sites it never cleared the bar regardless of which version it ran.
An early-warning instrument that can’t yet vouch for its own signal isn’t ready to vouch for anyone else’s. We’d rather publish that finding now than after the kit is already trusted.
— Dr Fenna Okoro, Senior Lecturer and Convenor, Horizon Register
Associate Professor Casimir Beng, Lead of the Adaptive Metrics Lab, welcomed the finding as consistent with the School’s wider work on measurement systems: “a reliability figure is worth publishing precisely because it’s the boring, unglamorous number nobody wants to be the one holding.”
The full poster is available from the University’s research repository under an open licence, doi:10.5555/slop.5xnve3.