Ethical metrics
Indicators that print their formula, alert thresholds and the refusal list.
Ethical metrics
The idea
A small set of team-level indicators where every card can open and show how it was calculated: the formula, the source events and the exclusions. If a number cannot show its working, it does not belong on the page.
What is on the page
- Snapshot trend — stored snapshots over time, so a metric can be read as a direction rather than a single value.
- Workload balance per member — open load distribution across the team.
- Priority-fairness matrix — who is carrying the urgent and high-priority work, computed from /api/tasks rather than from a separate store.
- Load vs seniority band — whether the heaviest load is landing on the least experienced people.
- Alert thresholds & limits — the configuration that decides when an indicator raises an alert.
Configuring it
- Open the configure dialog on /dashboard/ethical-metrics.
- Set thresholds and limits. The draft is stored through /api/ethical-metrics/config.
- Save. Config changes apply to the next snapshot.
Applying a recommendation
When an indicator crosses a threshold the page offers a recommendation. Accepting it calls POST /api/ethical-metrics/apply-recommendation, which performs the underlying change (for example reassigning load) and records it — it does not silently edit a metric.
Unavailable state
If the supporting data is missing, the page shows an "Ethical metrics unavailable" card explaining what is absent instead of rendering zeros. Zeros and missing data are different claims.
The refusal list
Published alongside the indicators:
- No lines of code.
- No per-person velocity.
- No activity, presence or keyboard-time scores.
- No ranking of people against each other by default.
Snapshots
Each snapshot is a row you can list and delete through /api/ethical-metrics and /api/ethical-metrics/<id>. Deleting a snapshot removes it from the trend on the next render.