[GH#260] [Future] Evaluate optional SQLite JSON1 / FTS5 extensions for local search and read-model queries #26

New issue

Open

opened 2026-05-19 22:15:45 +02:00 by Max · 0 comments

Max commented

2026-05-19 22:15:45 +02:00

Owner

Migrated from GitHub #260
Originally created by @Bio1988 on 2026-04-23T09:42:58Z

Context

Post-MVP follow-up from the SQLite/storage audit.

SQLite is currently a good fit as the local system of record for Strategy Desktop. However, some optional SQLite extensions may become useful later for local search and richer read-model queries.

Why this is future work

This is not needed for the MVP recorder/replay path. The current priorities are storage correctness, retention, session semantics, and bounded-context cleanup.

Question

Which SQLite extensions, if any, are actually worth enabling later in this product?

Initial candidates worth evaluating:

JSON1 for more ergonomic querying of JSON-backed read models and payloads
FTS5 for local full-text search across briefings, logs, notes, or derived text artifacts

Non-goals

do not add extensions just because they exist
do not optimize raw telemetry ingestion with extensions that do not materially improve the hot path
do not complicate deployment unless there is a clear user-facing benefit

Scope

document whether the current SQLite driver already exposes the needed functionality
identify the product surfaces that would benefit from JSON1 or FTS5
evaluate deployment, portability, and migration impact on Windows desktop packaging
recommend adopt / defer / reject for each extension

Acceptance criteria

recommendation exists for JSON1
recommendation exists for FTS5
rationale is grounded in actual Strategy Desktop read/query use cases
if a recommendation changes architecture or packaging assumptions, add an ADR

Notes

This is explicitly a post-MVP feature exploration ticket for the Future Features bucket.

Migrated from [GitHub #260](https://github.com/Bio1988/strategy-desktop/issues/260) Originally created by @Bio1988 on 2026-04-23T09:42:58Z --- ## Context Post-MVP follow-up from the SQLite/storage audit. SQLite is currently a good fit as the local system of record for Strategy Desktop. However, some optional SQLite extensions may become useful later for local search and richer read-model queries. ## Why this is future work This is not needed for the MVP recorder/replay path. The current priorities are storage correctness, retention, session semantics, and bounded-context cleanup. ## Question Which SQLite extensions, if any, are actually worth enabling later in this product? Initial candidates worth evaluating: - `JSON1` for more ergonomic querying of JSON-backed read models and payloads - `FTS5` for local full-text search across briefings, logs, notes, or derived text artifacts ## Non-goals - do not add extensions just because they exist - do not optimize raw telemetry ingestion with extensions that do not materially improve the hot path - do not complicate deployment unless there is a clear user-facing benefit ## Scope - document whether the current SQLite driver already exposes the needed functionality - identify the product surfaces that would benefit from `JSON1` or `FTS5` - evaluate deployment, portability, and migration impact on Windows desktop packaging - recommend adopt / defer / reject for each extension ## Acceptance criteria - recommendation exists for `JSON1` - recommendation exists for `FTS5` - rationale is grounded in actual Strategy Desktop read/query use cases - if a recommendation changes architecture or packaging assumptions, add an ADR ## Notes This is explicitly a post-MVP feature exploration ticket for the `Future Features` bucket.

Max added the

labels

2026-05-19 22:15:45 +02:00