RSOS: an executable spatial-symbolic language for human–AI and AI–AI communication
RSOS is a programming language whose programs are spatial and symbolic compositions (objects, places, flows, beams, colours, dots, rooms, senses) and whose operators (heal, beam, contain, compress, inherit, transplant, re-host, fuse, pierce, route by state) act on any host: a person, a team, an AI agent, an organisation or a physical system. It is a compiled language with 108 statements, governance checked at compile time (consent, standing consent in networks with laws, a seven-virtue Spine), a runtime, a two-way painting interface, a communication layer for compressed, self-healing messages, a keyboard layer, and the White Dragon, a cross-domain design engine over spheres. Version 2 reports pre-registered tests on real LLM agents (about 11,800 model calls) and the first human test. A hologram palace protocol designed by the White Dragon used 0.55× the BPE channel tokens of a Slipstream-style anchor protocol while keeping every copy of the shared state true in 16/16 tasks (anchor protocol: 0/16) and healing every injected drift without a model call. The White Dragon also redesigned a weak agent system, cutting errors by 63% with its predicted gain within 16%, while blind changes made it worse. Of 27 pre-registered predictions on real hosts, 18 passed and 9 failed; every failure is reported. RSOS grew from one seed: healing, transforming, signalling and evolving.
Authors
- R. A. Elu
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-17
- DOI
- https://doi.org/10.5281/zenodo.24304403
- Primary Topic
- Logic, programming, and type systems
- Type
- preprint