The Dataset Friction Framework: anonymised user surveys and aggregated support-ticket analysis (supplementary data)

Supporting data for Pidduck, E. & Modigliani, U., The Dataset Friction Framework: measuring user-facing friction as a complement to FAIR (submitted to Scientific Data). This deposit contains the data underlying the survey- and ticket-based results in the paper:• Two anonymised ECMWF user surveys — the Open Data Survey 2024 (N = 581) and the Licensed User Survey 2025 (N = 68) — each with a README, response table, and codebook.• Aggregated, non-identifying outputs of the 18,556-ticket support-corpus analysis: dimension-level counts, a stratified temporal sample, and the classification taxonomy. The raw support-ticket corpus is not included: it comprises third-party personal data collected during ECMWF operational support, for which there is no consent or legal basis for redistribution under the EU GDPR, and reliable de-identification of free-text records at this scale is not feasible. The aggregated outputs provided here underlie every ticket-based figure and statistic in the paper. Survey responses were anonymised: direct identifiers were removed, organisation names in the licensed survey were replaced by a coarse sector category, and free text was screened and redacted.