{"id":4,"date":"2026-02-24T15:50:28","date_gmt":"2026-02-24T15:50:28","guid":{"rendered":"https:\/\/www.w3.org\/community\/pm-kr\/?page_id=4"},"modified":"2026-02-28T17:14:48","modified_gmt":"2026-02-28T17:14:48","slug":"procedural-memory-knowledge-representation-pm-kr-community-group","status":"publish","type":"page","link":"https:\/\/www.w3.org\/community\/pm-kr\/procedural-memory-knowledge-representation-pm-kr-community-group\/","title":{"rendered":"Procedural Memory Knowledge Representation (PM-KR) Community Group"},"content":{"rendered":"\n

## Mission<\/strong><\/p>\n\n\n\n

The PM-KR Community Group develops standards for **procedural knowledge representation**<\/strong> that enable both humans and AI systems to consume the same canonical knowledge sources.<\/p>\n\n\n\n

**Core insight (Milton Ponson, mandala graph theory):**<\/strong> Nothing is “wrong” with declarative approaches \u2014 they’re **necessary but insufficient**<\/strong>. PM-KR provides **procedural optimization given declarative foundation**<\/strong>.<\/p>\n\n\n\n

**Boundary framework (Christoph Dorn):**<\/strong> Reality is not uniform, containing paradoxes and non-logical choices. PM-KR addresses hard\/soft\/blurred\/broken boundaries at fractal levels, with structural transparency as the safety net for author accountability.<\/p>\n\n\n\n

### Triple Foundation<\/strong><\/p>\n\n\n\n

PM-KR provides **procedural knowledge representation**<\/strong> where:<\/p>\n\n\n\n

– **Declarative foundation**<\/strong> (Milton Ponson): Semantics, structure, mathematical rigor (mandala graph theory)<\/p>\n\n\n\n

– **Procedural execution**<\/strong> (PM-KR): Runnable, renderable, multi-modal<\/p>\n\n\n\n

– **Boundary framework**<\/strong> (Christoph Dorn): Hard\/soft\/blurred\/broken boundaries at fractal levels<\/p>\n\n\n\n

– **Structural transparency**<\/strong> (Christoph Dorn): Author accountability, safety net for AI systems<\/p>\n\n\n\n

– **Reality modeling**<\/strong> (Christoph Dorn): Captures paradoxes, non-logical choices, ignorance (not uniform models)<\/p>\n\n\n\n

**The result:**<\/strong> AI systems that are:<\/p>\n\n\n\n

1. **Mathematically rigorous**<\/strong> (Milton’s mandala graph theory)<\/p>\n\n\n\n

2. **Philosophically grounded**<\/strong> (Christoph’s boundary framework)<\/p>\n\n\n\n

3. **Practically implementable**<\/strong> (reference implementations)<\/p>\n\n\n\n

4. **Ethically accountable**<\/strong> (structural transparency = safety net)<\/p>\n\n\n\n

—<\/p>\n\n\n\n

## Problem Statement<\/strong><\/p>\n\n\n\n

Current knowledge representation systems suffer from massive duplication and fragmentation:<\/p>\n\n\n\n

– The same knowledge is duplicated across fonts, embeddings, accessibility metadata, and visual renderings<\/p>\n\n\n\n

– Human-readable and machine-readable formats diverge, requiring separate maintenance<\/p>\n\n\n\n

– AI training data duplicates knowledge already available in structured forms<\/p>\n\n\n\n

– Knowledge updates must be propagated across multiple systems manually<\/p>\n\n\n\n

**Result:**<\/strong> Inefficiency, inconsistency, and unsustainability at scale.<\/p>\n\n\n\n

—<\/p>\n\n\n\n

## Approach: Declarative Foundation + Procedural Execution<\/strong><\/p>\n\n\n\n

**Declarative approaches**<\/strong> (RDF, OWL, JSON-LD) provide the **”know-what”**<\/strong> \u2014 foundational understanding (structure, relationships, semantics).<\/p>\n\n\n\n

**PM-KR adds the “know-how”**<\/strong> \u2014 procedural execution layer that makes declarative knowledge **runnable, renderable, and multi-modal**<\/strong>.<\/p>\n\n\n\n

**We’re not replacing RDF\/OWL\/JSON-LD \u2014 we’re adding the execution layer on top.**<\/strong><\/p>\n\n\n\n

### Core Principles<\/strong><\/p>\n\n\n\n

1. **Dual-Client Contract**<\/strong>: One canonical procedural source serves both humans (readable) and AI systems (executable)<\/p>\n\n\n\n

2. **Compositional Architecture**<\/strong>: Knowledge units compose via symlink-style references (no duplication)<\/p>\n\n\n\n

3. **Procedural Foundation**<\/strong>: Knowledge is executable programs (like TrueType fonts, mathematical formulas, physics simulations)<\/p>\n\n\n\n

4. **Hyper-Modularity**<\/strong>: Atomic knowledge units that combine into complex structures<\/p>\n\n\n\n

5. **Explicit Context Rules**<\/strong>: Context-dependent meanings handled via inspectable procedural rules (not implicit neural weights)<\/p>\n\n\n\n

### Synergy with Declarative Standards<\/strong><\/p>\n\n\n\n

| **Approach**<\/strong> | **Role**<\/strong> | **Execution**<\/strong> | **Transparency**<\/strong> |<\/p>\n\n\n\n

|————–|———-|—————|——————|<\/p>\n\n\n\n

| **Declarative (RDF\/OWL)**<\/strong> | **Foundation**<\/strong> (“know-what”) | \u274c No | \u2705 Yes |<\/p>\n\n\n\n

| **PM-KR (Procedural)**<\/strong> | **Execution layer**<\/strong> (“know-how”) | \u2705 Yes | \u2705 Yes |<\/p>\n\n\n\n

| **Declarative + PM-KR**<\/strong> | **Complete system**<\/strong> | \u2705 Yes | \u2705 Yes |<\/p>\n\n\n\n

—<\/p>\n\n\n\n

## Scope<\/strong><\/p>\n\n\n\n

### In Scope<\/strong><\/p>\n\n\n\n

**Core Standards:**<\/strong><\/p>\n\n\n\n

– Procedural knowledge data models<\/p>\n\n\n\n

– Composition semantics (symlink references, deduplication)<\/p>\n\n\n\n

– Execution semantics (RPN-based procedural programs)<\/p>\n\n\n\n

– Context rule semantics (handling context-dependent meanings)<\/p>\n\n\n\n

– Conformance levels (minimal, extended, sovereign)<\/p>\n\n\n\n

**Domains:**<\/strong><\/p>\n\n\n\n

– Mathematical knowledge (symbols, operators, formulas)<\/p>\n\n\n\n

– Spatial knowledge (geometric primitives, transformations)<\/p>\n\n\n\n

– Linguistic knowledge (characters, glyphs, typography)<\/p>\n\n\n\n

– Educational knowledge (textbooks, curricula, multi-modal rendering)<\/p>\n\n\n\n

– Game mechanics (rules, systems, procedural generation)<\/p>\n\n\n\n

– Scientific knowledge (protocols, experiments, simulations)<\/p>\n\n\n\n

**Relationships with W3C Technologies:**<\/strong><\/p>\n\n\n\n

– JSON-LD (structured data representation)<\/p>\n\n\n\n

– RDF\/OWL (semantic web integration \u2014 PM-KR adds execution layer)<\/p>\n\n\n\n

– Verifiable Credentials (knowledge provenance)<\/p>\n\n\n\n

– CBOR-LD (compression for efficient transmission)<\/p>\n\n\n\n

### Out of Scope<\/strong><\/p>\n\n\n\n

– Proprietary AI training formats (focus: open, standardized)<\/p>\n\n\n\n

– Natural language understanding (we provide structured knowledge inputs)<\/p>\n\n\n\n

– Inference engines (we define knowledge representation, not reasoning systems)<\/p>\n\n\n\n

—<\/p>\n\n\n\n

## Deliverables (2026-2027)<\/strong><\/p>\n\n\n\n

### Specifications<\/strong><\/p>\n\n\n\n

1. **PM-KR Core Specification v1.0**<\/strong><\/p>\n\n\n\n

– Procedural knowledge data model<\/p>\n\n\n\n

– Composition semantics<\/p>\n\n\n\n

– Conformance levels<\/p>\n\n\n\n

– **Target:**<\/strong> Q4 2026<\/p>\n\n\n\n

2. **PM-KR Execution Semantics v1.0**<\/strong><\/p>\n\n\n\n

– RPN-based procedural programs<\/p>\n\n\n\n

– Interpreter requirements<\/p>\n\n\n\n

– Sandbox\/security model<\/p>\n\n\n\n

– **Target:**<\/strong> Q1 2027<\/p>\n\n\n\n

3. **PM-KR Context Rules Specification v1.0**<\/strong><\/p>\n\n\n\n

– Context-dependent execution semantics<\/p>\n\n\n\n

– Inheritance and override rules<\/p>\n\n\n\n

– Multi-modal rendering patterns<\/p>\n\n\n\n

– **Target:**<\/strong> Q2 2027<\/p>\n\n\n\n

4. **PM-KR JSON-LD Profile v1.0**<\/strong><\/p>\n\n\n\n

– JSON-LD context definitions<\/p>\n\n\n\n

– Vocabulary terms<\/p>\n\n\n\n

– Canonical serialization rules<\/p>\n\n\n\n

– **Target:**<\/strong> Q2 2027<\/p>\n\n\n\n

### Reference Implementation<\/strong><\/p>\n\n\n\n

– **Knowledge3D**<\/strong> (Python\/CUDA): Sovereign spatial procedural knowledge system with GPU execution<\/p>\n\n\n\n

– Demonstrates: Hyper-modular architecture, 50-90% compression, dual-client contract<\/p>\n\n\n\n

### Use Case Documentation<\/strong><\/p>\n\n\n\n

– **Education:**<\/strong> Procedural textbooks for human and AI tutors (MIT OpenCourseWare example)<\/p>\n\n\n\n

– **Gaming:**<\/strong> Executable rulebooks for game masters and AI assistants (D&D SRD example)<\/p>\n\n\n\n

– **Science:**<\/strong> Procedural experimental protocols (Nature Protocols example)<\/p>\n\n\n\n

– **Accessibility:**<\/strong> Multi-modal knowledge rendering (visual, audio, tactile from ONE source)<\/p>\n\n\n\n

– **AI Training:**<\/strong> Canonical knowledge sources (Wikipedia procedural KB \u2014 query, don’t duplicate)<\/p>\n\n\n\n

—<\/p>\n\n\n\n

## Current Members (18+ as of Feb 28, 2026)<\/strong><\/p>\n\n\n\n

**Chairs:**<\/strong><\/p>\n\n\n\n

– Daniel Campos Ramos (Founder, Knowledge3D Project)<\/p>\n\n\n\n

– Milton Ponson (Co-Chair, Mathematical Foundations)<\/p>\n\n\n\n

**Notable Organizations:**<\/strong><\/p>\n\n\n\n

– MIT Digital Credentials Consortium<\/p>\n\n\n\n

– Huawei Technologies (W3C Advisory Board)<\/p>\n\n\n\n

– Digital Bazaar (Manu Sporny, JSON-LD co-creator)<\/p>\n\n\n\n

– LinkedIn (Knowledge Graphs)<\/p>\n\n\n\n

– University of Brescia, Italy<\/p>\n\n\n\n

– Indiana University<\/p>\n\n\n\n

– Rensselaer Polytechnic Institute<\/p>\n\n\n\n

– INRIA, France<\/p>\n\n\n\n

– Cogsonomy<\/p>\n\n\n\n

—<\/p>\n\n\n\n

## Liaisons with Other W3C Groups<\/strong><\/p>\n\n\n\n

**Active Collaborations:**<\/strong><\/p>\n\n\n\n

– Verifiable Credentials WG: Knowledge provenance and attribution<\/p>\n\n\n\n

– JSON-LD WG: Procedural JSON-LD extensions<\/p>\n\n\n\n

– Spatial Data on the Web WG: Spatial knowledge representation<\/p>\n\n\n\n

– Credentials CG: Academic credentials and structured educational knowledge<\/p>\n\n\n\n

**Potential Future Liaisons:**<\/strong><\/p>\n\n\n\n

– Semantic Web Interest Group<\/p>\n\n\n\n

– Web Machine Learning WG<\/p>\n\n\n\n

– RDF-star WG (nested knowledge structures)<\/p>\n\n\n\n

—<\/p>\n\n\n\n

## Roadmap<\/strong><\/p>\n\n\n\n

**Q1 2026 (Current):**<\/strong><\/p>\n\n\n\n

– \u2705 Community Group launched (Feb 20, 2026)<\/p>\n\n\n\n

– \u2705 18+ members recruited<\/p>\n\n\n\n

– \u2705 Mission statement revised based on community feedback (v1.0 \u2192 v1.2)<\/p>\n\n\n\n

– \ud83d\udd04 Initial specification drafts (PM-KR Core v0.1)<\/p>\n\n\n\n

**Q2 2026:**<\/strong><\/p>\n\n\n\n

– Publish PM-KR Core Specification v0.5 (draft for community review)<\/p>\n\n\n\n

– Gather community feedback (use cases, implementations)<\/p>\n\n\n\n

– Establish liaisons with related W3C WGs<\/p>\n\n\n\n

– Expand reference implementations (JavaScript, Rust)<\/p>\n\n\n\n

**Q3 2026:**<\/strong><\/p>\n\n\n\n

– Publish PM-KR Core Specification v1.0 (candidate)<\/p>\n\n\n\n

– Host W3C TPAC breakout session<\/p>\n\n\n\n

– Empirical validation studies (compression, performance, accuracy)<\/p>\n\n\n\n

**Q4 2026:**<\/strong><\/p>\n\n\n\n

– Finalize PM-KR Core Specification v1.0<\/p>\n\n\n\n

– Begin PM-KR Execution Semantics v1.0 draft<\/p>\n\n\n\n

– Publish conformance test suite<\/p>\n\n\n\n

**2027:**<\/strong><\/p>\n\n\n\n

– PM-KR Execution Semantics v1.0<\/p>\n\n\n\n

– PM-KR Context Rules Specification v1.0<\/p>\n\n\n\n

– PM-KR JSON-LD Profile v1.0<\/p>\n\n\n\n

– Industry adoption (textbook publishers, game companies, AI platforms)<\/p>\n\n\n\n

—<\/p>\n\n\n\n

## Why PM-KR Matters Now<\/strong><\/p>\n\n\n\n

### Technical Impact<\/strong><\/p>\n\n\n\n

– **Standardize**<\/strong> procedural knowledge representation for AI systems<\/p>\n\n\n\n

– **Reduce**<\/strong> knowledge duplication across systems (50-90% compression demonstrated)<\/p>\n\n\n\n

– **Enable**<\/strong> dual-client knowledge sources (humans + AI from same source)<\/p>\n\n\n\n

– **Complement**<\/strong> declarative standards (add execution layer to RDF\/OWL\/JSON-LD)<\/p>\n\n\n\n

### Societal Impact<\/strong><\/p>\n\n\n\n

– **Education:**<\/strong> Accessible, multi-modal textbooks (visual, audio, tactile)<\/p>\n\n\n\n

– **Sustainability:**<\/strong> Reduce AI’s carbon footprint via compression (no training data duplication)<\/p>\n\n\n\n

– **Accessibility:**<\/strong> Knowledge rendered for diverse human needs (same source)<\/p>\n\n\n\n

– **Reproducibility:**<\/strong> Scientific protocols as executable procedures<\/p>\n\n\n\n

### Economic Impact<\/strong><\/p>\n\n\n\n

– **Efficiency:**<\/strong> Companies stop duplicating knowledge infrastructure<\/p>\n\n\n\n

– **Innovation:**<\/strong> New applications enabled by compositional knowledge<\/p>\n\n\n\n

– **Open Standards:**<\/strong> Prevent proprietary lock-in, foster ecosystem growth<\/p>\n\n\n\n

—<\/p>\n\n\n\n

## How to Participate<\/strong><\/p>\n\n\n\n

**Join:**<\/strong> https:\/\/www.w3.org\/community\/pm-kr\/ (no W3C membership required)<\/p>\n\n\n\n

**Contribute:**<\/strong><\/p>\n\n\n\n

– Review specifications in GitHub: https:\/\/github.com\/danielcamposramos\/Knowledge3D\/tree\/main\/docs\/vocabulary<\/p>\n\n\n\n

– Propose use cases from your domain<\/p>\n\n\n\n

– Build prototypes in your preferred language<\/p>\n\n\n\n

– Participate in discussions: public-pm-kr@w3.org<\/p>\n\n\n\n

—<\/p>\n\n\n\n

## Contact<\/strong><\/p>\n\n\n\n

**Chairs:**<\/strong><\/p>\n\n\n\n

– Daniel Campos Ramos: capitain_jack@yahoo.com<\/p>\n\n\n\n

– Milton Ponson: rwiciamsd@gmail.com<\/p>\n\n\n\n

**Mailing List:**<\/strong> public-pm-kr@w3.org<\/p>\n\n\n\n

**GitHub:**<\/strong> https:\/\/github.com\/danielcamposramos\/Knowledge3D<\/p>\n\n\n\n

—<\/p>\n\n\n\n

## Acknowledgments<\/strong><\/p>\n\n\n\n

PM-KR builds on decades of W3C work in semantic web, linked data, and web standards.<\/p>\n\n\n\n

**Special thanks:**<\/strong><\/p>\n\n\n\n

– **Dave Raggett**<\/strong> (W3C veteran) \u2014 Critical feedback on declarative vs procedural positioning<\/p>\n\n\n\n

– **Milton Ponson**<\/strong> (Co-Chair) \u2014 Mathematical insight: declarative foundation + procedural optimization synergy (mandala graph theory)<\/p>\n\n\n\n

– **Christoph Dorn**<\/strong> (Systems Architect) \u2014 Boundary framework: hard\/soft\/blurred\/broken boundaries, structural transparency as safety net for author accountability<\/p>\n\n\n\n

– **Manu Sporny**<\/strong> (JSON-LD co-creator) \u2014 JSON-LD integration guidance<\/p>\n\n\n\n

– **Tim Berners-Lee**<\/strong> \u2014 Linked Data principles<\/p>\n\n\n\n

– All 18+ founding members who joined in the first week<\/p>\n\n\n\n

—<\/p>\n\n\n\n

**Version:**<\/strong> 1.3 (Triple Foundation: Milton + Christoph + Implementation)<\/p>\n\n\n\n

**Last Updated:**<\/strong> February 28, 2026<\/p>\n","protected":false},"excerpt":{"rendered":"

## Mission The PM-KR Community Group develops standards for **procedural knowledge representation** that enable both humans and AI systems to consume the same canonical knowledge sources. **Core insight (Milton Ponson, mandala graph theory):** Nothing is “wrong” with declarative approaches \u2014 … Continue reading →<\/span><\/a><\/p>\n","protected":false},"author":24487,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"_s2mail":"yes","footnotes":""},"class_list":["post-4","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/www.w3.org\/community\/pm-kr\/wp-json\/wp\/v2\/pages\/4","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.w3.org\/community\/pm-kr\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/www.w3.org\/community\/pm-kr\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/www.w3.org\/community\/pm-kr\/wp-json\/wp\/v2\/users\/24487"}],"replies":[{"embeddable":true,"href":"https:\/\/www.w3.org\/community\/pm-kr\/wp-json\/wp\/v2\/comments?post=4"}],"version-history":[{"count":4,"href":"https:\/\/www.w3.org\/community\/pm-kr\/wp-json\/wp\/v2\/pages\/4\/revisions"}],"predecessor-version":[{"id":11,"href":"https:\/\/www.w3.org\/community\/pm-kr\/wp-json\/wp\/v2\/pages\/4\/revisions\/11"}],"wp:attachment":[{"href":"https:\/\/www.w3.org\/community\/pm-kr\/wp-json\/wp\/v2\/media?parent=4"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}