The Building Speaks Before You Shiver

Expectation. The building shall detect failures in its own infrastructure and inform affected residents before they discover the failure themselves.

Required.

• When a building system that affects residents fails — elevator, water supply, climate control, electrical supply, or any other shared infrastructure — the building communicates the failure to affected residents through the primary interface. The communication includes what failed, the affected area, and the expected duration or next update time.
• The building does not rely on residents reporting an infrastructure failure as the primary means of detection. Where the building monitors a system, a failure in that system is communicated to affected residents without waiting for the first complaint.
• No outage communication persists in the interface for longer than 48 hours without an update. A notice that reads "temporarily unavailable" on day one and still reads "temporarily unavailable" on day fourteen has failed. The building updates the status or removes the notice.
• When an infrastructure failure is resolved, the building communicates the resolution to the same residents who were notified of the failure. No resident continues to avoid an elevator, carry water, or reroute their day based on a failure that was resolved hours ago.
• The building publishes which infrastructure systems it monitors and which it does not. A resident can see where the building has detection capability and where it relies on resident reports or manual inspection.
• When the building's monitoring, notification, or communication platform is changed, replaced, or upgraded, the published environmental standards, notification configuration, and outage communication history are preserved without gap or loss ACC.RE +2.

Recommended.

• Where the building monitors infrastructure systems, the building detects a failure before the first resident reports it. The resident's first interaction with the failure is the notification, not the experience.
• When a failure is detected that will predictably affect a specific subset of residents — a floor, a wing, a riser — the notification is targeted to affected residents, not broadcast building-wide.

In practice.

A resident wakes up on a Tuesday morning. Before they step into the shower, a notification appears: hot water supply interrupted — boiler fault detected at 5:15 AM, affecting floors 8–14, repair crew dispatched, estimated restoration by 10 AM. The resident adjusts their morning. They did not discover the outage by standing under cold water.

The elevator in tower B stops at 7:30 AM. The building's monitoring system detects the fault. By 7:35 AM, residents of tower B receive a notification: elevator B1 out of service — fault detected, technician en route, estimated restoration 2 hours, elevator B2 operating normally. The resident on floor 22 takes elevator B2 instead of waiting at a door that will not open.

The pool pump is repaired after three days of closure. A notification reaches every resident who received the original closure notice: pool reopened — pump repaired, water treated, normal operating hours resumed. The resident who had stopped checking because the pool was closed learns it is open again.

A new resident checks the building's monitoring disclosure in the interface: elevators — monitored, water supply — monitored, climate system — monitored, air quality in common areas — monitored, parking ventilation — not monitored. The resident knows where the building has detection capability. The building did not claim to monitor everything; it disclosed what it covers.

The building migrates its notification platform to a new vendor. Published environmental standards — comfort ranges, maintenance schedules, air quality thresholds — transfer to the new system. The outage history is preserved. The configuration that targets floor-specific notifications survives. The migration did not reset the building's communication infrastructure to a blank state.

Failure modes.

Cold shower discovery. No hot water. No notification. No notice in the lobby. The resident steps into a cold shower. They email the office. No response until the afternoon. The building knew about the boiler fault — the BMS logged it at 5 AM — but no one translated the system alert into a resident notification.

Eternal temporary. The elevator has been out of service for two weeks. A paper sign on the door reads "temporarily out of order." No date. No estimated restoration. No updates. The word "temporarily" has lost all meaning. Residents have stopped expecting the elevator to return and have started planning their days around the remaining one.

Resolved but unknown. The water supply was restored at 2 PM. No notification was sent. Residents who left for the day to avoid the outage return in the evening and check the interface: the outage notice is still active. The failure was resolved six hours ago. The resident's avoidance behavior continued because the resolution was not communicated.

Floor 3 hears about floor 14. The building sends a building-wide notification about a water shutoff affecting floors 12–16. Residents on floor 3 read the notification, check their taps, find everything working, and discard the message. The next notification they receive — about something that actually affects them — gets less attention because the last one was irrelevant.

Test.

1. Simulate an infrastructure failure in a monitored system. Confirm: affected residents are notified through the primary interface with what failed, the affected area, and expected duration or next update time.
2. Resolve the failure. Confirm: the same residents receive a resolution notification.
3. Review all active outage notices in the interface. Confirm: none has persisted longer than 48 hours without an update.
4. Review the last five infrastructure failures. Confirm: the building's notification preceded the first resident complaint in each case where the system was monitored.
5. View the building's monitoring disclosure. Confirm: it lists which infrastructure systems are monitored and which are not.
6. Migrate the notification or monitoring platform. Confirm: published standards, notification configuration, and outage history are preserved.