Research Status / P-Programming

Where P-programming stands

Short answer: the primitives are real, the discipline is not complete. The UM runner and papers now support event spaces, LPPs, shift-chain memory, conjunction-style threshold creation, SN persistence, and some surprise-conditioned structure. What is still missing is a finished, general theory of how larger P-programs are composed and how higher-level structures are realized without imperative stand-ins.

Done Core primitives are codified in the spec and realized in the runner.
Partial Lexicon-scale designs exist, but some experiments still use hash tables or other stand-ins.
Missing Canonical composition rules and a complete higher-level P-programming grammar.

Staged Checklist

Primitive substrate

Codified. #umr_spec and #working_memory_paper fix the base machinery: ESs, LPPs, shift chains, transient clearing, conjunction products, threshold creation, and SN as interchange format.

Runner realization

Implemented. The runner supports the low-level operations that real P-programs need, including dynamic threshold-created contexts and some conditional structures such as #umr_cond_lpp.

Lexicon-oriented program templates

Partially codified. #lexicon_embedding_paper, #trigram_embedding_paper, and #lexicon_experiments describe the main bytes-to-bigrams-to-words path, but they are still partly design documents rather than completed canonical programs.

End-to-end higher-level examples

Mixed. The system has working examples, but some key experiments still declare their own P-programming gaps. The clearest case is word discovery, where word identity is still tracked by a hash table rather than fully by UM-native reification.

Composition laws

Not yet codified. There is no final document that says, in general form, how complex P-programs compose, how they are staged, and what the canonical legal constructions are above the primitive layer.

General programming discipline

Not yet complete. P-programming is past the “idea only” stage, but not yet at the point where a new contributor could treat it as a finished programming system with a settled grammar and settled design laws.

Next Codification Moves

  1. Write the canonical composition rules for multi-stage P-programs.
  2. Replace imperative stand-ins in lexicon work with threshold-created UM-native structure.
  3. Promote one lexicon-scale path to the first end-to-end reference P-program.
  4. Decide what belongs in SN directly versus what remains runner dynamics.
  5. Document the finished discipline as a short reference paper and a build-visible overview page.

Current evidence base:

  • #umr_spec is the canonical reference for the current mathematical substrate.
  • #mcp_p_programs_analysis explicitly marks the 20260215 P-programs paper as operational, not a source of new mathematical laws.
  • #lexicon_embedding_feedback records that an explicit component P-programs section was removed until implementation exists.
  • #word_discovery_paper and #umr_word_discover explicitly record the current gap.