The idea, and why it's fair
Communities around character-creation tools — from open avatar makers to games whose studios have since closed — have produced millions of characters. Each one is a point in that tool's parameter space: a vector of choices about proportion, feature, and form. Taken together, that's a map of what people actually build and find appealing.
We want to learn the shape of that map to steer our own procedural generator — but we do not want, and do not take, anyone's meshes, textures, or files. The distinction is the whole program:
- Facts and statistics aren't copyrightable. Measuring a distribution — how often a choice recurs, where the space is dense or empty — needs no license, because it copies nothing.
- We retain zero asset bytes. An asset is measured and discarded; only numbers survive, and a gate enforces it mechanically rather than by promise.
- We give the measurement back. The aggregate analysis is published openly, so the contributors whose collective work we studied can use it to see where their community's space is crowded and where it's wide open — and grow their own work into the gaps.
That last point is the reciprocity. An open analysis of a community's collective output is a gift back to that community. It's the CC-attribution spirit made mutual: we learn from the aggregate, and the aggregate learns from itself through our lens.
The pipeline — one substrate, many adapters
Every source, however different, reduces to the same shape: a parser that turns it into rows of measured facts, feeding one shared study stage. A rigged model, a parametric character card, and an academic body dataset all land in the same normalized frame.
The phases
Phase 0 — the substrate foundation
A single measured-facts store (one row per source-and-fact, with provenance and license tags), and the extraction contract with a gate that asserts zero asset bytes survive. The honest guarantee is mechanical, not a claim.Phase 1 — source adapters in progress
One parser per corpus shape: rigged glTF/VRM models (mesh, skeleton, material, UV census); parametric character cards (the parameter block is the data — no mesh involved); and open academic body datasets folded in as ground-truth anchors. Each adapter emits only numbers.Phase 2 — distribution studies the data product
Per source: a taxonomy of parameters; distributions over the community's vectors (what recurs, what clusters, where coverage is dense or sparse); and a normalization that maps every tool's space onto one common frame so they can be compared at all. The output is a coverage report with a number on each gap.Phase 3 — generator wiring the private win
The measured distributions become target distributions for our own generator, which must first cover the community's measured span, then exceed it on axes a fixed asset library can't follow — per-character regeneration, physically invariant motion, no authored files.Phase 4 — the give-back the point
Each study is published as an open brief with attribution to the source community, and its data points — the taxonomies and distributions — released as open data. This page is the first of them: the methodology itself, given back before any specific study.Guardrails — mechanical, not promised
- Zero asset bytes retained. Enforced by a gate, so it can't quietly drift.
- Aggregate only. A distribution is published only when it's drawn from enough independent sources to describe no individual's work — a statistic, never a reconstruction.
- One license check, at one gate. Only publishing someone's actual bytes needs a license (open-licensed excerpts, with credit). Measurements need none, because they copy nothing.
- Attribution always names the source community.
UNVERIFIED — what this page does not yet establish
- This is the methodology and roadmap, not a completed study. No distribution has been published yet; the phases past the substrate are in progress or planned.
- The reciprocity claim — that contributors find the aggregate analysis useful — is a stated intent, not a measured outcome. It becomes real only when the first data-point release is used.
- The zero-bytes and k-anonymity guarantees are described here; their enforcing gates are to be built and shown, not taken on trust.
- Nothing here endorses redistributing any source's assets. The program is designed specifically to make that unnecessary.
Sources & foundations
- BEDLAM 2.0 — 8M+ synthetic bodies in SMPL-X, an open academic anchor for anthropometric ground truth
- AGORA — multi-person synthetic dataset with SMPL/SMPL-X fits, a second open ground-truth spine
- MakeHuman — open-source parametric human generator; its CC0 output is a clean, fully-usable base
- CC BY 4.0 — the attribution model this program mirrors and extends into reciprocity
- Character surface floor — the measure-don't-ship pattern applied to textures
- Rig floor — the same pattern applied to skeletons
Lineage — Forks off the surface-floor and rig-floor measurements, which established that a reference corpus can be measured into a capability target without taking its assets. Parent: character asset floor, 2026-08-03.