starforged-std-3001-appendix-e

Population, demographic and anthropometric data sets for Starforged players and characters, modeled after NASA-STD-3001 Vol 2 Appendix E.

What it is for

The data sets describe the physical accommodation span for Starforged systems. They follow the structure of NASA-STD-3001 Vol 2 Appendix E, widened from a U.S. Army population to the worldwide civilian population, because Starforged is played worldwide. The human config holds the rows for real Earth players. The starforged config holds the rows for the fictional setting population. The conversion config maps between the two in both directions. Every row has the same columns. universe tells the two populations apart, and section, category, metric, value, percentile, sex, population, source and notes carry each value with its citation. The Markdown sources are APPENDIX-E.md for the human universe and APPENDIX-E-STARFORGED.md for the setting. GUIDELINES.md is the translation guide from a real player to a Starforged character. QA.lean encodes the distributions in Lean 4 and proves invariants over them, such as distribution sums, percentile monotonicity and cross-universe section coverage.

Provenance

• Modeled after NASA-STD-3001 Vol 2 Appendix E.
• Sex, age and regional composition: UN World Population Prospects 2024 Revision and UN DESA 2024.
• Anthropometric anchor: ISO 7250-1:2017 methodology with DINED (TU Delft) aggregation. ANSUR is replaced here.
• Range of motion and strength: AAOS civilian orthopedic reference (Norkin and White), Bohannon 2019, NIOSH RWL and Mathiowetz 1985.
• Handedness: Papadatou-Pastou et al. 2020.
• Intersex estimate: Sax 2002.
• Setting population: the Nem and Mila Social VR Lifestyle Survey 2024 (Bredikhina and Nem).
• Full citation metadata is in CITATION.cff.

Use

from datasets import load_dataset
human = load_dataset("chibifire/starforged-std-3001-appendix-e", "human")
starforged = load_dataset("chibifire/starforged-std-3001-appendix-e", "starforged")
conversion = load_dataset("chibifire/starforged-std-3001-appendix-e", "conversion")

Check the proofs with lean QA.lean. It needs only Lean 4 core, not Mathlib.

Licence

CC-BY-4.0. See the Creative Commons Attribution 4.0 International License. Cited data sets keep their own licences. UN WPP and UN DESA data is public. ANSUR I is a U.S. government work in the public domain. Papadatou-Pastou et al. 2020 is CC-BY-4.0.