Closing the Execution Gap: A Policy Framework for Independent Verification of Computational Research

This project examines the structural reasons why reproducibility failures persist even under ideal Registered Reports (RR) conditions and outlines the system‑level requirements for independently verifying computational execution. While RR have transformed scientific publishing by ensuring transparency of intent, they do not verify execution. Modern computational workflows are complex, fragile, and often opaque to peer review; reviewers cannot rerun analyses, reconstruct environments, or detect silent deviations. These challenges are infrastructural rather than editorial. The materials in this project present a policy‑oriented framework for addressing the “execution gap” through research infrastructure that supports independent re‑execution, tamper‑evident audit trails, controlled data domains, and validator independence. The framework emphasises privacy, locality, and GDPR alignment: sensitive data remains under the researcher’s control, and only digital fingerprints and high‑level summaries are shared. The goal is to support international discussion among funders, institutions, journals, and research‑integrity bodies about how execution‑verification infrastructure should be assembled and governed. This project forms part of a broader conceptual trilogy exploring: the execution gap, the architectural requirements for independent verification, and the governance principles needed to ensure trustworthiness and resistance to institutional capture. All materials are conceptual and policy‑focused. They do not represent production code or an operational system.