The Story
A 93-year-old woman enters a building that cannot complete her profile.

A Multimodal Literary Analysis
When a smart building tries to understand a tenant who leaves almost no digital trace.
System status
Project Overview
A 93-year-old woman enters a building that cannot complete her profile.
Elevator-5 can collect fragments of her life but cannot fully understand her.
A Marxist reading of ownership, data, control, and human value.
Smart buildings are designed to anticipate needs, optimize comfort, and simplify daily life. Through interconnected systems, they create environments that respond, adapt, and care, making every moment seamless and efficient.
22°C
Optimal
Stable
All systems normal
Secure
No threats detected
On Track
7:00 12:00 18:00 22:00
From sensors, logs, and connected services, the building collects patterns that help it make predictions and recommendations.
Basic
Available
Last update: 3 days ago
24
Transactions
Last 30 days
Mostly
Home
Low external travel
Classical
Mostly
Low variety detected
Routine
Stable
Confidence: Medium
The new resident arrives without the synchronized data the building expects. Elevator-5 detects only fragments of her daily life, while her destination, routine, and preferences remain largely unknown.
Detected fragments:
refrigerator contents, thermostat setting, washing-machine cycles.
Unregistered
Tap to hear
(Chung)
No routine
Tap to hear
(Chung)
Preference unknown
Tap to hear
(Chung)
Unsynced
Tap to hear
(Chung)
Limited access
Tap to hear
(Chung)
Elevator-5 can read the data, but it cannot understand the human behind it.
Who she is

vs.
What the system can see
Profile confidence: low
The building expects complete data to predict and serve.
But some lives cannot be measured, only witnessed.
Technology in “A Song for Sleep” is not neutral. It reflects economic structures, extracts value from daily life, and determines whose humanity is recognized.
The building’s owners control the system, while residents have little control over how their lives are monitored.
Private information becomes a resource used for profiling, prediction, and commercial benefit.
Behavioral patterns allow the system to anticipate residents’ needs and influence future consumption.
The system measures residents through data but cannot measure memory, care, grief, or dignity.