The Taxonomy

A shared vocabulary for describing technology.

Each category is a question people ask about a technology, and its elements are the answers. Choose a category, then an element, to see its icon and what it means. New to the terms? Read the concepts.

Collection Technology

Elements 13

What captures the data this system uses?

The devices and channels that capture this system's input data — cameras, sensors, microphones, forms, or feeds from other systems. The instrument that collects the data, distinct from the algorithm that processes it.

About this category

Collection Technology

What captures the data this system uses?

Datachain

Answer: Only systems that capture data in the world use a device for it. One that works from existing records has none.
Icon shape: The system itself: why it exists, who is involved, what it does, and the points where it meets the world.

Variables

Description Optional

Biometric ScannerCaptures a physical trait for identification — fingerprint, palm, iris, or a face scan at a gate, kiosk, or turnstile. The capture device itself; matching the scan against stored templates is biometric recognition, which appears under Algorithm or Model.

CameraCaptures still images or video of a place — mounted on a pole, a vehicle, a drone, or built into a device. Includes infrared and thermal cameras. This is the capture device itself; the techniques that analyze the footage appear under Algorithm or Model, and how identifiable the footage is appears on the Input Data category.

Connected Device TelemetryReadings streamed automatically by connected things — vehicles reporting speed and position, wearables reporting heart rate, smart meters reporting energy use, machines reporting their own status. Continuous capture that runs in the background of everyday use.

Environmental SensorMeasures conditions in the surroundings — air quality, temperature, humidity, noise level, water level, or weather. These sensors read the environment, not people, though noise and air readings can still reflect human activity.

Existing RecordsData the organization already holds — customer accounts, case files, registries, transaction histories, past applications. Nothing new is captured from the world; the system draws on what was recorded before, sometimes for a different original purpose.

MicrophoneCaptures sound — speech, ambient noise, or specific audio events like breaking glass or gunshots. Found in smart speakers, kiosks, intercoms, and acoustic monitoring installations. The capture device itself; speech recognition and other analysis appear under Algorithm or Model.

Network ConnectionConnects this system to a network — a cellular modem, a WiFi access point, or a wired link. It moves data rather than capturing it, but connection logs can still reveal which devices were nearby or connected, and an access point that offers public WiFi is also a service people join.

Positioning & LocationDetermines where something is — GPS receivers in vehicles or apps, Bluetooth beacons, or indoor positioning systems. Produces location traces, one of the most revealing kinds of personal data when tied to a person or their device.

Presence & Motion SensorDetects that someone or something is there, or is moving, without capturing an image — passive infrared, ultrasonic, pressure mats, door counters, seat occupancy sensors. Tells the system that a space is occupied, not who occupies it.

Third-Party Data FeedData received from another organization — purchased, licensed, or delivered through an API or data-sharing agreement. Examples: credit bureau records, weather services, map providers, data brokers. The people the data describes may never have dealt with this system's operator.

User InputInformation people provide deliberately — a form they fill in, text they type, a search they run, a photo they upload, a voice command they speak. The person knows they are handing something over, unlike sensing that happens around them.

Vehicle DetectionSenses vehicles on a road or in a space — radar, LiDAR, induction loops in the pavement, piezoelectric strips, or gantry-mounted detectors. Used to count traffic, measure speed, and detect congestion. License-plate cameras belong under Camera.

Wireless Signal SensingDetects the radio signals devices give off — WiFi probes from phones, Bluetooth, RFID tags, or NFC taps. Often used to count or track devices moving through a space; a device's signal can act as a persistent identifier for the person carrying it.

Biometric Scanner

Captures a physical trait for identification — fingerprint, palm, iris, or a face scan at a gate, kiosk, or turnstile. The capture device itself; matching the scan against stored templates is biometric recognition, which appears under Algorithm or Model.

Variables

Description Optional