Skip to content

CHANGELOG.md

Latest commit

 

History

History
189 lines (146 loc) · 8.31 KB

CHANGELOG.md

File metadata and controls

189 lines (146 loc) · 8.31 KB

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

Unreleased

Added

  • [Schema]: EntityData field for ontology-typed memories

    • New entity property with name (required), entity_type, and entity_id fields
    • Supports additional properties defined by ontology entity_type schemas
    • Links structured data to ontology definitions

  • [Schema]: Block references field

    • New blocks object for named block references (^block-id)
    • Maps block identifiers to their text content for granular linking

  • [Schema]: Shared EntityReference definition

    • Extracted to schema/definitions/entity-reference.schema.json
    • Reused by both MIF schema and Citation schema
    • Prevents definition divergence across schemas

  • [Schema]: Ontology extends field for inheritance

    • Ontologies can declare parent ontologies to inherit from
    • Enables trait inheritance model: mif-base → shared-traits → domain
    • Added to ontology.schema.json and all domain ontologies

  • [Project]: VERSION.json for centralized version constants

    • Specification version, schema versions, ontology versions
    • Single source of truth for all version numbers

Changed

  • [Schema]: Standardized schema identifiers to https://mif-spec.dev/schema/ namespace

    • Updated ontology.schema.json $id
    • Updated all domain ontology schema_url fields
    • Note: These are identifiers, not resolvable URLs

  • [Schema]: Discovery patterns structure updated

    • Split single patterns[] into content_patterns[] and file_patterns[]
    • Each pattern type has specialized fields for its use case
    • Aligns schema with actual mif-base ontology usage

  • [Examples]: Updated JSON-LD context property names

    • Changed dc:created to created, dc:modified to modified
    • Consistent with unprefixed field names in specification

  • [Examples]: Clarified Level 1 conformance requirements

    • namespace is recommended but optional at Level 1
    • Level 1 only requires: id, type, created, and content body

Fixed

  • [Examples]: Fixed ontology reference format in example memories

    • Changed from incorrect ontology.entity_type to proper ontology.id + entity block
    • Updated 6 example memory files across agriculture, publishing, biology-lab directories

  • [Documentation]: Added trait inheritance documentation to ontologies/README.md

    • Documents the three-tier trait system
    • Explains how domain ontologies inherit from shared-traits

  • [Documentation]: Added decay model rationale (Section 9.3)

    • Explains P7D/P14D/P30D half-life values as pragmatic defaults
    • Documents scientific background from Ebbinghaus forgetting curve
    • Cites Murre & Dros (2015), Squire & Bayley (2007), Wickelgren (1972)
    • Provides guidance for tuning values per memory type

Changed

  • [Schema]: Memory types now use three base types
    • Replaced ad-hoc types (memory, decision, preference, fact, episode, pattern, learning, context)
    • New base types: semantic (facts/knowledge), episodic (events/experiences), procedural (processes/how-to)
    • Specific categorization expressed through namespace hierarchy (e.g., _semantic/decisions, _episodic/incidents)
    • Ontologies can extend types via entity_types with base field
    • BREAKING: Existing memories using old type values need migration

Added

  • [Schema]: OntologyReference field in MIF schema

    • New ontology property for declaring which ontology a memory applies
    • Fields: id (required), version (optional), uri (optional)
    • Enables validation that memories conform to declared ontology
    • Links MIF documents to ontology definitions

  • [Ontology]: Industry-specific ontology examples

    • regenerative-agriculture.ontology.yaml - Farm operations, carbon credits, certifications
    • k12-educational-publishing.ontology.yaml - K-12 curriculum, state adoptions
    • biology-research-lab.ontology.yaml - Academic research, grants, compliance
    • backstage.ontology.yaml - Developer portal entity catalog
    • shared-traits.ontology.yaml - Reusable trait mixins

  • [Backstage]: Backstage.io catalog integration examples

    • Example catalog-info.yaml files for each industry ontology
    • MIF-to-Backstage entity mapping via annotations

  • [Ontology]: Three-type namespace hierarchy

    • Base ontology with semantic/episodic/procedural top-level namespaces
    • Nine sub-namespaces: decisions, knowledge, entities, incidents, sessions, blockers, runbooks, patterns, migrations
    • Entity type definitions with traits and schemas
    • Relationship types with cardinality constraints
    • Discovery patterns for content and file-based detection

  • [Ontology]: JSON-LD semantic web support

    • ontology.context.jsonld for semantic vocabulary mapping
    • yaml2jsonld.py converter script
    • Alignment with Schema.org and SKOS vocabularies

  • [Schema]: Ontology validation schema

    • schema/ontology/ontology.schema.json for YAML validation
    • Supports hierarchical namespace children
    • Entity type and relationship validation

  • [Specification]: Initial MIF (Memory Interchange Format) specification

    • JSON-LD based format for AI memory interoperability
    • Bi-temporal model with valid time and transaction time
    • W3C PROV-compliant provenance tracking
    • Conformance levels (Core, Extended, Full)
    • Human-readable Markdown export support

  • [Specification]: Citations structure (Level 3 optional feature)

    • Structured citation references with type/role taxonomy
    • Required fields: type, title, url, role
    • Optional fields: author, date, accessed, relevance, note
    • Entity references in author field using wiki-link syntax
    • Citation types: article, book, paper, website, documentation, repository, video, podcast, specification, dataset, tool
    • Citation roles: supports, refutes, background, methodology, contradicts, extends, derived, source, example, review
    • Frontmatter YAML schema and body section Markdown syntax
    • JSON-LD vocabulary with Schema.org alignment
    • Validation rules (Section 5.5.7) with field constraints and error handling
    • Appendix D: Citations Quick Reference

  • [Specification]: Compression fields (Level 3 optional feature)

    • summary - Concise 2-3 sentence summary (max 500 characters)
    • compressed_at - Timestamp when compression was applied
    • Compression criteria: Age > 30 days AND lines > 100, OR Strength < 0.3 AND lines > 100

  • [Schema]: JSON Schema for automated validation

    • schema/mif.schema.json - Complete MIF document validation
    • schema/citation.schema.json - Standalone citation object validation
    • Draft 2020-12 compliant schemas with comprehensive type definitions

  • [Examples]: Reference MIF document examples

    • Basic memory interchange examples
    • Entity and relationship examples
    • Temporal metadata examples
    • Level 3 citations example (level-3-citations.memory.md/.json)

Changed

  • [README]: Updated to reflect new features

    • Added Citations and JSON Schema to Key Features table
    • Added Validation section with schema usage examples
    • Updated Level 3 conformance to include citations and compression
    • Updated examples description

  • [CONTRIBUTING]: Added JSON Schema validation guidance

Documentation

  • [Research]: Comprehensive market research report

    • Competitive landscape analysis (Mem0, Zep, Letta, LangMem, Cognee, Graphlit)
    • Standards alignment review (JSON-LD, RDF/OWL, ONNX, PROV)
    • Enterprise requirements assessment (EU AI Act, GDPR, NIST AI RMF)
    • Adoption strategy recommendations

  • [Research]: Executive brief for decision support

    • Market opportunity summary ($2.1B SAM)
    • Competitive positioning analysis
    • Prioritized action items

  • [Research]: Trend models and forecasting

    • Market growth projections (2024-2030)
    • Technology adoption S-curve
    • Scenario analysis (standard adoption vs fragmentation)
    • Regulatory impact timeline (EU AI Act milestones)

0.1.0 - 2026-01-23

Added

  • Initial project setup
  • MIF specification draft v0.1
  • Market research framework