GLC Metadata Builder

Getting started with GLC metadata

Make light-exposure and optical-radiation datasets easier to understand, validate, and reuse.

The Global Light Commons (GLC) standard provides a shared structure for describing studies, participants, devices, datasheets, and measurement files for light-exposure and optical-radiation research. This builder guides you through creating a complete metadata package that can accompany your dataset.

Note: files you select while building the metadata package stay on your computer. They are processed locally in your browser and are not uploaded, transmitted, or stored by this site.

Why researchers use GLC

GLC metadata makes wearable light-exposure and optical-radiation datasets easier to review, combine, and reuse.

Easier collaboration Study, participant, device, and dataset information can be found in predictable places.

Validation-ready Links between files can be checked automatically to catch common mistakes.

Reusable data Future users can understand what was measured, with which device, and under what conditions.

What you’ll create

A GLC package links the key parts of a wearable light-exposure or optical-radiation dataset.

Study Participants Devices & Datasheets Dataset Files

The resulting metadata package can be validated and shared alongside your data.

How the builder works

Work through the pages in order, import existing metadata when available, then export a package for validation.

Describe your study
Add participants
Document devices and datasheets
Register dataset files
Export a package

Expected time: 5–30 minutes.

Generated files and package structure

A GLC metadata package keeps package-level metadata at the root and resource metadata in data/.

glc-metadata-package/
  datapackage.json
  data/
    study.json
    participants.csv
    participant_characteristics.csv
    devices.json
    device_datasheet.json
    datasets.json
    datasets/
      your-measurement-files.csv

study.jsonWhat was the study, who contributed, and which dataset records belong to it?

participants.csvWho or what was measured?

participant_characteristics.csvWhich optional participant-level characteristics help interpret the measurements?

devices.jsonWhich devices were used and which datasheets describe them?

device_datasheet.jsonWhat sensor, calibration, channel, or spectral details are needed to interpret the device output?

datasets.jsonWhere are the measurement files, what columns do they contain, and how do they link to study, participant, and device IDs?

Useful resources

Learn more about the Global Light Commons initiative View human-readable details of the GLC datapackage structure

1

Project setup

Choose the release version and package-level settings.

Already have metadata?

Import an existing package folder to review or continue editing it here.

The builder looks for datapackage.json, study.json, participants.csv, participant_characteristics.csv, devices.json, device_datasheet.json, and datasets.json at the folder root or inside data/.

Schema version * Package name Package title

2

Study

Core fields from study.schema.json. Dataset IDs will be filled from the dataset step.

Optional: import an existing study.json to pre-fill this page.

Study internal ID * Study title * Pre-registration Registration Ethics Study type Short description * Study sample * Study setting * Geographical location * Intervention Funding sources Keywords All study datasets * generated from Datasets page

Study groups

Optional repeatable groups from study_groups. Select the dataset records that belong to each group.

Study contributors

Optional embedded contributors from contributor.schema.json. Add rows manually, or import a CSV/TSV/JSON file to fill the table.

Import replaces the contributor rows below. For CSV/TSV, use these column names: contributor_full_name, contributor_roles, contributor_email, contributor_orcid, contributor_institution_name, contributor_institution_city, contributor_institution_country. Roles can be separated with semicolons. JSON can be either contributors.json or study.json with study_contributors.

Download metadata:

Reference schemas:

3

Participants

Add participant rows for participants.csv and optional characteristics.

Participant roster

Required rows for participants.csv.

Optional: import participants.csv or participants.json. Expected fields: participant_internal_id, participant_age, participant_sex, participant_gender.

Participant characteristics

Optional long-format rows for participant_characteristics.csv.

Optional: import participant_characteristics.csv or JSON. Expected fields: participant_internal_id, participant_characteristic_name, participant_characteristic_value, participant_characteristic_unit, participant_characteristic_description.

Download metadata:

Reference schemas:

4

Devices

Add device records and the datasheets they reference.

Device inventory

Required records for devices.json.

Optional: import devices.json or a CSV/TSV device table. Expected fields: device_internal_id, device_manufacturer, device_model, device_serial_number, device_calibration_date, device_firmware_version, device_datasheet_id, device_sensors. Sensor entries can use sensor type | sensor datasheet id, separated by semicolons.

Datasheet IDs are pending cross-references: if the matching datasheet has not been added yet, leave the field blank or keep the imported ID and complete the datasheet section below.

Device datasheets

Add device or sensor datasheet records referenced by devices.

Optional: import device_datasheet.json, sensor_datasheet.json, or a CSV/TSV datasheet table. Expected fields include datasheet_id, datasheet_manufacturer, datasheet_type, datasheet_model, datasheet_calibration_interval, spectral sensitivity, calibration fields, and channels.

Download metadata:

Reference schemas:

5

Dataset record ?

One dataset record represents one participant dataset. Cross-reference fields use earlier pages.

Dataset or file group?

Use a new dataset record for a different participant, primary device, body/device location, or primary modality. Use a file group for files that belong to the same participant measurement context, including auxiliary files such as wear logs.

Example: wrist light data and its wear log can be one dataset with two file groups. Wrist light and chest light for the same participant should usually be two dataset records.

Optional: import an existing datasets.json. Multiple dataset records are imported as separate participant-level records.

Dataset records ?

Each record should describe one participant dataset. Add more records when a study has multiple participants or participant-level datasets.

Dataset ID * Study ID * Participant ID * Device ID * Device location * Sampling interval * Dataset timezone * Dataset location: latitude * Dataset location: longitude *

Dataset instructions *

5a

Dataset file assistant ?

Each file group describes one set of same-structure files for this participant.

No file selected yet.

File format * Encoding * Header row File timezone * Auxiliary file * Preprocessing applied Preprocessing description

Variable terms for this file group

These terms populate the semantic-term dropdowns for variables in the active file group. other is always included.

Datetime for this file group

Use column metadata when the file contains date/time columns. If it does not, declare the collection date/time instead.

Datetime source * Choose whether this file group has date/time columns or only a known collection date/time. Collection date/time value Use this only when the file has no date/time column. Date or datetime column Column containing a full datetime, or the date column if date and time are separate. Date/datetime format * Format of the selected date or datetime value. Separate time column Only choose a column here if date and time are stored in two separate columns. Time format Use this to describe the separate time column format, for example HH:mm:ss.

5b

Detected variables

Edit labels, units, calibration notes, semantic terms, and tick primary variables.

Select a file to detect columns.

Download metadata:

Reference schemas:

6

Export package

Download individual metadata files or a metadata package zip.

The zip includes generated metadata files and schemas. It does not include the original data files referenced by datasets.json.