Capture
Camera-first personal discards and litter observations with AI interpretation, confidence and source provenance.
THE 2018 PRODUCT, REBUILT AROUND EVIDENCE
the whole thing, again.
YUK is not a recycling scanner bolted onto old software. The modern implementation is the convergence point for World Cleanup mapping, cleanup coordination, open data and machine waste recognition.
Sites
Persistent physical places that accumulate evidence instead of disposable trashpoint pins.
Cleanup
Interventions and verification change believed state without deleting the history of what happened.
Offline
Client-generated IDs and an idempotency ledger establish the contract for durable mobile operation logs.
Open data
Privacy-safe views expose rounded geometry and derived classifications while originals stay private.
Organizations
Community groups, municipalities and cleanup crews are first-class entities instead of identity hacks.
Disposal rules
Authoritative local rules have their own versioned data layer rather than being invented by the model.
AI
Models interpret sensors, suggest matches and classify evidence. Human corrections supersede output instead of erasing it.
Imports
World Cleanup, Open Data, WADE, TrashAI and future datasets enter as attributed sources and observations.
capture → observation → site → verification → intervention → current belief
Personal discards can stay private. Litter can become public evidence. Cleanup adds an intervention. Reappearance adds another observation. Nothing important needs to be destroyed to make the interface look current.