Slate v2 Perfect Architecture Research Plan

Goal

Use the normalized raw corpora for Lexical, ProseMirror, Tiptap, and Tiptap Docs to produce a deeper architecture/API analysis for the Slate v2 perfect plan:

• what to steal
• what to reject
• what to defer

Research Questions

1. What lifecycle and mutation boundaries should Slate v2 expose publicly?
2. Which extension/plugin ideas are worth stealing directly?
3. Which engine/runtime ideas should stay internal only?
4. Which field ideas should be rejected because they conflict with a data-model-first React-perfect runtime?
5. Which ideas are valuable but should be deferred to a later batch?

Phases

1. Re-read compiled architecture coverage and identify gaps.
2. Read strongest raw architecture/API hits per corpus.
3. Write deeper source summaries where existing pages are too narrow.
4. Write a system map for steal/reject/defer.
5. Write one or more decision pages for the perfect Slate v2 plan.
6. Update index/log and close the corpus ledger.

Status

• Phase 1: complete.
• Phase 2: complete.
• Phase 3: complete.
• Phase 4: complete.
• Phase 5: complete.

Outputs

Added:

• docs/research/systems/slate-v2-perfect-plan-steal-reject-defer-map.md
• docs/research/decisions/slate-v2-perfect-plan-should-steal-read-update-transaction-discipline-and-extension-dx.md

Updated:

• docs/research/index.md
• docs/research/log.md
• docs/research/systems/editor-architecture-landscape.md

Corpus Closure

• Lexical: evidenced
• ProseMirror: evidenced
• Tiptap/Tiptap Docs: evidenced