p235

This repository contains collections of design patterns and organizational metaphors with formal specifications, including a virtual hardware implementation (NPU-253) for accelerated pattern operations.

🚀 NPU-253: Neural Processing Unit / Natural Patterning Unit

NEW: Virtual hardware device implementing the 253-pattern language as a memory-mapped coprocessor!

from npu253 import PatternCoprocessorDriver, NPUConfig

npu = PatternCoprocessorDriver(NPUConfig())
npu.load()

# Query patterns
pattern = npu.query_by_id(1)
print(f"{pattern.name}: {pattern.solution}")

# Domain transformation
social = npu.transform_pattern("12610010", "social")

Features:

• 🔧 Hardware-style MMIO register interface
• 🎯 253 APL patterns + 253 archetypal patterns
• 🔄 Domain transformation (physical/social/conceptual/psychic)
• ⚡ LRU caching for performance
• 📊 Telemetry and diagnostics
• ✅ 34 passing tests

Documentation:

• NPU-253 Blueprint - Architecture and design
• NPU-253 API - Complete API reference
• npu253/README.md - Package documentation
• Demo | Tests

# Run demo
python3 demo_npu253.py

# Run tests
python3 test_npu253.py

📖 Documentation

Formal Specifications

• Formal Specification Summary - Complete overview of architecture and Z++ specifications
• docs/ - Comprehensive technical documentation (2,328 lines)
  • Architecture overview with Mermaid diagrams
  • Z++ formal specifications (data model, system state, operations, integrations)
  • Usage guide for specifications

Pattern Collections

• APL: Christopher Alexander's "A Pattern Language" (253 patterns) - architectural and urban design patterns
• UIA: Union of International Associations "Patterns & Metaphors" (253 patterns) - organizational and conceptual patterns

Markdown Versions

The markdown/ directory contains converted versions of all APL and UIA pages in markdown format for improved readability and accessibility:

• markdown/apl/: 253 A Pattern Language patterns (apl001.md - apl253.md)
• markdown/uia/: 253 UIA Patterns & Metaphors (12610010.md - 12612530.md)
• markdown/arc/: 102 Archetypal Patterns extracted from UIA templates with domain-specific placeholders

See markdown/README.md for detailed information about the conversion process and markdown structure.

Pattern Language Schema

The repository includes formalized JSON schemas for both the Pattern Language and Archetypal Patterns:

APL Pattern Language Schema

• Pattern Language Schema (pattern_language_generated.json) - Complete meta-pattern, categories, and sequences
• Pattern Sequences (pattern_sequences.json) - All 36 pattern sequences with emergent phenomena
• Categories (category_*.json) - Towns, Buildings, and Construction categories

See PATTERN_SCHEMA_README.md for detailed information about the APL schema structure.

Archetypal Pattern Schema

• Archetypal Pattern Schema (archetypal_pattern_schema.json) - JSON schema for archetypal patterns
• Archetypal Patterns (archetypal_patterns.json) - All 102 archetypal patterns with domain mappings
• Placeholder Reference (archetypal_placeholders.json) - Complete placeholder documentation

Archetypal patterns use the format: "generic {{domain-specific}} generic" and can be transformed across:

• Physical - Spatial, material, architectural domains
• Social - Organizational, community, institutional domains
• Conceptual - Knowledge, theoretical, paradigmatic domains
• Psychic - Awareness, consciousness, mental domains

See ARCHETYPAL_SCHEMA_README.md for detailed information about the archetypal pattern schema.

Generating Schemas

# Generate APL Pattern Language schema
python3 generate_pattern_schema.py

# Generate Archetypal Pattern schema
python3 generate_archetypal_schema.py

Testing Schemas

# Test APL schema (if tests exist)
python3 validate_schema.py

# Test Archetypal schema
python3 test_archetypal_schema.py

Demo Schemas

# Demo APL schema
python3 demo_pattern_schema.py

# Demo Archetypal schema
python3 demo_archetypal_schema.py

OpenCog Atomese Representation

The Pattern Language has been converted to OpenCog's Atomese format for knowledge representation and pattern matching:

• opencog_atomese/ - Complete Atomese hypergraph representation in Scheme format
  • pattern_language.scm - Complete representation
  • meta_pattern.scm - Meta-pattern
  • categories.scm - Categories with InheritanceLinks
  • sequences.scm - Sequences with MemberLinks
  • Enhanced Features:
    • pattern_language_enhanced.scm - Enhanced with diagrams, details, and connections
    • relationship_types.scm - Pattern relationship types (complement, conflict, alternative)
    • patterns/ - Individual .scm files for modular loading

The Atomese format enables:

• Pattern matching and reasoning in OpenCog
• Hypergraph queries for pattern relationships
• Knowledge graph navigation and inference
• Integration with AI/AGI systems

See opencog_atomese/README.md for usage examples and opencog_atomese/ENHANCEMENTS.md for enhanced features documentation.

Paradigm & Language Analysis

Comprehensive analysis of optimal implementation approaches for achieving cognitive "optimal grip" on the gestalt salience landscape:

• PARADIGM_LANGUAGE_ANALYSIS.md - Complete paradigm and language evaluation

  • Analysis of 6 programming paradigms (hypergraph, functional, logic, OOP, constraint, agent-based)
  • Evaluation of 6+ languages (Scheme, Datalog, Python, Haskell, JavaScript, Prolog)
  • Cognitive affordances mapping
  • Recommended multi-layer architecture

• IMPLEMENTATION_GUIDE.md - Practical implementation guide

  • Concrete code examples for each layer
  • Datalog query system integration
  • Haskell domain transformation engine
  • Python salience computation
  • D3.js visualization framework

• PARADIGM_COMPARISON_MATRIX.md - Detailed comparison matrices

  • Cognitive requirements vs paradigms matrix
  • Implementation tasks vs languages matrix
  • Paradigm synergy analysis
  • Use-case specific recommendations

Key Finding: Multi-paradigm approach required - Hypergraph (Scheme/OpenCog) foundation + Datalog queries + Functional transformations (Haskell) + Python integration + JavaScript visualization provides optimal cognitive affordances.

Generating Atomese Files

# Generate base Atomese files
python3 generate_opencog_atomese.py

# Generate enhanced features
python3 generate_enhanced_atomese.py

Testing Atomese Files

# Test base files
python3 test_opencog_atomese.py

# Test enhanced features
python3 test_enhanced_atomese.py

Demo Enhanced Features

python3 demo_enhanced_atomese.py

Demo Query System

# Demo Datalog query integration (requires pyDatalog)
pip install pyDatalog
python3 demo_datalog_queries.py