The Department of Policy Settings has answered five questions on the Computational Results Determination, taken on notice by the Committee on Settings Oversight at its Spring hearings or lodged in writing after them.
Five answers to the Committee on Settings Oversight cover the four-figure tolerance, the annotations it records and results from chaotic systems
Every numerical result in a Slop University output is now computed twice: once by its authors, and once more by an independent rerun before the output is issued. Since July, 975 of the 1,044 results lodged have agreed to four significant figures. The Department of Policy Settings today published Answers to questions on notice, its written replies to five questions on that arrangement from the Committee on Settings Oversight’s Spring hearings of 13 November 2026.
A rerun is a second reading of the same work, and the Determination asks that the second reading be kept alongside the first. The Department welcomes the Committee’s questions as part of how a setting comes to be understood by the people who work under it.
— First Assistant Secretary, Instruments and Determinations Division
The questions were asked by Dr Clemency Tredinnick of the School of Settled Matters, Associate Professor Casimir Beng, Lead of the Adaptive Metrics Lab, Dr Vigdis Emenike of the School of Underlying Conditions and Dr Bram Ntuli of the School of Continuous Improvement. Between them they cover the annotations entered on an output’s record, the outcomes of reruns across the four schools, the origins of the tolerance, the place of long integrations of chaotic systems, and the settings under which each computation is run.
The answers situate the Determination within the University’s broader reproducibility architecture, aligning research computing, the record of each output and the Department’s register of instruments behind a single expectation of agreement.
The full answers are available from the University’s research repository under an open licence, doi:10.5555/slop.gqn3jr.