Repository navigation
Standing queries: share evaluations, debounce, then DD-backed deltas #133
Description
Each subscription re-runs the full query on every relevant write, diffing via canonical JSON strings (reevaluate.rs:63). Identical subscriptions are not shared, and re-evaluations compete with interactive lookups for the permit.
Short term: share results per (kg, query, snapshot version), debounce bursts, run at background priority, diff on tuple hashes.
Long term: a DD-backed StandingQuery.
Roadmap checkpoint — 2026-10-03
R3 shared evaluation is isolated in #175; it is keyed by canonical query/KG/rule generation and equivalent visibility, with collision-safe equality and a send-time revocation fence. Coalescing must fit the measured latency budget. R4 DD attachment (#186) and default arranged lookup adoption (#187) finish this broader issue; keep it open through R4.
Activity
feat(subscriptions): shared authorized standing evaluation (#133)
feat(protocol): share standing-query evaluations across identical sub…
jsam commentedon Oct 4, 2026
Moved to Later / deferred. Standing-query evaluation work beyond PR #254 and #238 (now in Engine production readiness) is post-readiness dataflow work.
The neurosymbolic decision layer plan narrows the roadmap to two milestones: Engine production readiness (current focus), then Gateway neural integration (Jev, CLM, Laya) (#266). This issue stays open and can be pulled back in. Tracker: #199.
jsam commentedon Oct 5, 2026
Superseded by #317 (V10) in the incremental-views plan, epic #305.