How a desk's signals are sourced

A Signal Desk desk is built from its operator's own data. Every signal shows the source record it stands on.

The data is the operator's

An operator delivers their own datasets to their desk: by API, from a storage bucket, by file drop or SFTP, or by upload. A desk states where each feed comes from: the supplier, the path it travelled, and its last delivery. Readers see that on the desk's own Data and provenance page.

What a signal is built from

• A source record. Each signal is built from a record in the operator's data and shows the record it stands on, so a reader can read one against the other.
• A watch section. A signal is filed under a section by meaning, so a reader finds it by subject.
• The desk's own data, and nothing else by default. An operator can choose to add DRM3's public record beside their own data. That is off unless they turn it on.

Where the data goes, and where it does not

A desk's data is served on that desk, to the people its operator invited. It does not enter DRM3 commons search, another desk, or a data pack.

A desk that adds the public record

When an operator adds DRM3's public record, those sources are open data. Each one is listed with its license in the DRM3 data-license registry. Where a source asks for attribution, we carry it.

Every signal is disclosed as AI-derived

Signal Desk derives its signals with AI. Every signal says so, and that label cannot be removed. A signal is a reading of a record. The source record is the authority: read it before you act on a signal.