From secrets and data chaos to open science — our (open‑ended) quest to publish data meaningfully

Researchers often find themselves navigating between raw, heterogeneous data and the intuitive knowledge that gives those data meaning. This talk shares a pragmatic journey from “secrets and data chaos” toward open, reusable research outputs, drawing on experiences from research support and real-world projects. We will examine our experiences why sharing data in a meaningful way can be difficult—giving honest insights on considerations around sensitivity and confidentiality, fragmented workflows, legacy formats, missing metadata, and tacit know-how—and how we have started to tackle these hurdles and transform them into drivers of quality and openness. The session outlines a lightweight, scalable approach to meaningful data publication grounded in the FAIR principles: defining scope early with data management plans; capturing context through README files, codebooks, and domain standards; ensuring provenance with versioning and workflow documentation; linking data, code, and methods via persistent identifiers; and choosing licenses and repositories that balance openness with compliance. We will discuss handling sensitive or proprietary components (tiered releases, anonymization, and synthetic data), and how small process changes—automated validation, templated metadata, notebooks—produce outsized gains in trust and reusability. Rather than a one-time “release,” I aim to give a perspective on open science as an iterative, community-engaged practice: publishing what is possible now, inviting feedback, and improving over time. Attendees will leave with concrete patterns and decision points to move their own datasets from private collections to interoperable, citable, and genuinely useful research assets.