Evidence acquires epistemic force not by remaining unchanged, but by surviving transformation. Leonelli’s account of data-intensive science relocates epistemology from the possession of supposedly raw facts to the organisation of their trajectories: world becomes trace, trace becomes dataset, dataset becomes model, and model enters new communities, institutions and public claims. At each passage, cleaning, classification, metadata, aggregation and visualisation modify what an evidential object can mean and which propositions it can legitimately sustain. Scientific validity must therefore be understood through epistemic scope rather than decontextualisation: provenance, intended use and interpretive environment delimit the claims for which transformed data remain reliable. Spatial research makes this especially explicit. A coordinate may appear universally portable, yet its evidential value depends upon resolution, temporal conditions, measurement practice and the scale at which comparison is attempted. Thus, neighbourhood observations cannot become metropolitan evidence merely through aggregation; scale itself is epistemically constitutive. Leonelli’s conception of research environments extends this principle to scientific organisation, demonstrating that institutions, instruments, standards, funding systems and collaborative arrangements actively determine what can be observed and combined. For Socioplastics, this insight becomes architectural. Open science thereby becomes not simply access, but an infrastructure for tracing evidential life. UncertaintyPassage names this grammar of bounded mobility: the capacity of evidence to travel productively while retaining explicit limits on what its transformations allow it to claim.
