Sprint 08

Table of Contents

This page documents a sprint (Sprint 08) of ORE Studio v0. It captures the sprint's mission, current status, and the stories that compose it. For the surrounding context — version goals, sprint order, and product identity — see Version 0.

Mission

Originally implement basic reference data. In practice the sprint expanded to cover the substrate beneath that work: database reproducibility, change-management infrastructure for amends/deletes, Qt UX polish across most dialogs, plus a full set of entity-creator skills to make the next entity cheaper.

Status

Field Value
State DONE
Parent version Version 0
Previous Sprint 07
Start 2025-12-30
End (expected) 2026-01-11
Now Sprint closed 2026-01-11. Three observability-review items and the accounts-dialog polish carry forward.
Waiting on None — carried-forward items are scheduled for sprint 09+.
Next Sprint 09
Release Notes Release notes
Last touched 2026-01-11

Achievements

  • Countries landed as the first proper reference-data entity end-to-end.
  • Database schema and seeding overhauled; TimescaleDB tier detection in place.
  • Native pagination adopted across list views (sqlgen 0.6.0).
  • Change management infrastructure landed: reason codes with UI.
  • Entity-creator skills suite extended to Qt, shell, HTTP, Wt, and CLI surfaces.
  • Wt/HTTP validated and documented as the second presentation tier.

Stories

For the definitions of the themes see Themes.

Infrastructure

Story State Start End Description
Database schema and seeding overhaul DONE   2025-12-31 rerun SQL from scratch with TimescaleDB tier detection; remove code seeders.
Comms substrate evolution DONE   2026-01-11 composable options, rename ores.shellores.comms.shell, dynamic channel discovery.
Native pagination support DONE   2026-01-07 adopt sqlgen 0.6.0.
Wt and HTTP review DONE   2026-01-06 validate the sprint-07 surfaces; document the HTTP service.
Change management infrastructure DONE   2026-01-11 reasons base + UI.
Observability follow-up BACKLOG     review pass only; three implementation tasks carry forward.

Product

Story State Start End Description
Qt UX polish DONE   2026-01-11 validate holiday features; assorted dialog fixes; log-viewer UI; accounts polish deferred.
Reference data: countries DONE   2026-01-11 first proper refdata entity, end-to-end.
Books domain modelling analysis DONE   2026-01-11 strategic next-modelling effort.

LLMs

Story State Start End Description
Entity-creator skills suite DONE   2026-01-11 Qt, shell, HTTP (incl. recipes), Wt, CLI.

Agile

Story State Start End Description
Sprint 08 housekeeping DONE   2026-01-11 backlog refinement, OCR scans; AI summary deferred.

Charts

Charts generated via sprint_charts cmake target.

PRs & Commits per Day

Dual-axis bar chart. PRs (left axis) and commits (right axis) per day. A high commits-to-PR ratio may indicate scope creep.

prs_commits.png

Daily Line Churn

Lines added (green) and deleted (red) per day. Building work produces mostly additions; refactoring produces a mix. Days with no churn may indicate blockers.

line_churn.png

Cumulative Stories Done

Line chart tracking stories marked DONE during the sprint. Steady upward slope is healthy; plateauing signals a stall.

stories_done.png

Retrospective

What went well

  • The database overhaul (license-tier detection, SQL populate centralisation) was a foundational piece that unlocked subsequent reference-data work; landed early and paid back through the rest of the sprint.
  • Sprint-07 follow-ups (Wt and HTTP review, observability review) arrived on schedule, even if some of the resulting fix work was deferred — separating review from fixes was the right call.
  • Six entity-creator skills landed in parallel; the countries story exercised the shell + Wt skills as their first real workload.

What hurt

  • Ambitious in volume: 11 stories / 31 tasks made it the biggest v0 sprint to date. The cost showed up as four deferred items (AI summary, accounts dialog, all three observability fixes).
  • ores.refdata doesn't have a versioned modelling doc with a stable ID; the "see also" linkage from the countries story has to point sideways at the skills story instead of at the component doc.
  • Change-management UI polish ran longer than expected, mostly because the icon / save-dialog / human-time pattern changes touched every existing dialog.

What changed

  • ores.shell is now ores.comms.shell — the name reflects its binary-protocol scope and frees the ores.shell slot for any future protocol-agnostic shell.
  • Seeding centralised in SQL populate scripts; C++ seeders gone.
  • All dialogs mark stale on external change rather than auto-reloading, a pattern that came out of the feature-flags work and was generalised across the codebase.
  • Native sqlgen pagination replaces the offset workarounds and the single_connection aggregation hack.

Emacs 29.1 (Org mode 9.6.6)