Release Planning & Document History

This section details how the business and software requirements of the Fast Layered Universal Document Engine (Flude) are distributed across developmental release milestones, and tracks the version history of this specification document.

Document Version History

We track the revisions of these specifications using a structured versioning schema:

Doc Version	Date	Phase Description	Lead Author	Status
0.7	2026-06-28	Requirements audit & v2.0 roadmap finalization	pavel.sokolov	Draft (Active)
0.6	2026-06-13	Testing and bug fixing	Sir Derryk	Approved (Superceded)
0.5	2026-06-13	Testing documentation	Sir Derryk	Approved (Superceded)
0.4	2026-06-11	Developing documentation	Sir Derryk	Approved (Superceded)
0.3	2026-06-09	Developing prototype	Sir Derryk	Approved (Superceded)
0.2	2026-06-08	Planning MVP	Sir Derryk	Approved (Superceded)
0.1	2026-06-07	Requirements gathering	Sir Derryk	Approved (Superceded)

• Version 0.7 Scope: Full requirements audit of the implemented codebase against the design specification. Gap analysis matrix finalized (GAPs 01–32). v1.0 MVP requirements authorised. v2.0 "AI-Accelerated Core & Quality Upgrade" roadmap scoped (10 items across 4 clusters). v3.0+ navigation architecture and language parity work formally deferred. Design docs updated to reflect actual v1.0 state.
• Version 0.6 Scope: Comprehensive software testing, regression bug fixing, and implementation of automated navigation structures in the Flude code generator (eliminating pageless sidebar categories and generating dedicated namespace pages).
• Version 0.5 Scope: Comprehensive verification and testing of compiled documentation portals, sanitization of internal development metadata, and final deployment quality assurance.
• Version 0.4 Scope: Comprehensive design, planning, and deployment of the hybrid online documentation portal (Flude Portal), including the GHA CI/CD publication workflow (Flude Publisher).
• Version 0.3 Scope: Active core implementation of Python-based Flude prototype according to TDD methodologies (Weeks 1 to 5).
• Version 0.2 Scope: Granular scheduling of 12 TDD tasks across 5 weeks, preparing executable development task specifications, and updating Docusaurus navigation layouts.
• Version 0.1 Scope: Gathering high-level business goals (BRD), defining system functional/non-functional constraints (SRS), drafting the pipeline design (SDD), and mapping out the implementation schedule.

---

Release Planning Summary

The implementation of Flude is structured into three release tiers, each building on a fully delivered predecessor:

Corrected 2026-08-10. "Navigation Architecture" (GAP-06/GAP-26 — sidebar.toml as the single navigation source, static/inline/redirect node rendering) was originally scoped to the v3.0+ tier below, but landed during v2.0 development instead (IMP-32.10, 2026-08-02) — see future_v2.md's "Breaking Changes from v1.0" table for detail. Removed from the v3.0+ section; the v3.0+ tier now covers only the items that are still genuinely undelivered (language render parity, AI enrichment, and the rest of future_v2.md's deferred roadmap).

1. MVP (v1.0) Release Plan: Delivered. Core parsing (Doxygen XML → IR), CommonMark normalization, HTML and Hugo Markdown rendering for C++, C#, Java, and Python, 3-way config cascade, fault-tolerance, L1 parse cache, and a 23-test unit and functional test suite.
2. v2.0 — AI-Accelerated Core & Quality Upgrade: 10 items across 4 clusters — Infrastructure modernisation (global config, logging, L2 cache, Doxyfile key-level merge), Library API & CLI unification (UdeOrchestrator public API, subcommands), Typed IR (7 Pydantic entity models), and QA completeness (coverage gate, external script confirmation, per-language integration suites) — plus Navigation Architecture (sidebar.toml as the mandatory single navigation source, static/inline/redirect node rendering), pulled forward from the v3.0+ tier during development.
3. v3.0+ — Language Parity & AI: Language render parity, Docomatic ODA HTML compatibility, fully static document generation, LLM docstring enrichment, multi-language translation, RAG/XML output formats, and native language parsers.