FAQ

Where do I put an application e-mail address?

Not in the feed. The feed is crawled and redistributed, so contact data in it ends up in third-party indexes. Put application instructions on the job page referenced by url. That's also why the schema has no field for them.

We have no open positions. Should we still publish a feed?

Yes. Publish "jobs": []. The domain stays healthy in SourceTracker, and consumers can tell "no openings" apart from "no feed".

Can a job url point to a different domain, e.g. our ATS?

Yes. The url may be on any domain, e.g. https://acme.jobs.example-ats.com/123. Only the feed itself must be discoverable on your domain.

We have several brands or subsidiaries on different domains.

Publish one feed per domain. Each feed describes the employer behind that domain.

Can I add custom fields?

No. additionalProperties is false for the feed, employer and job, so unknown fields fail validation. Propose new fields in GitHub Discussions.

How is this different from schema.org JobPosting?

JobPosting is markup embedded in a single job page. To find it, a crawler has to visit every page of the site. oJobPub lists all openings in one document at a fixed URL and links to those pages. Use both. See Relation to other formats.

The older json-job schema has been inactive for more than ten years and doesn't define a discovery location.

How do I get a list of all publishing domains?

Download the daily export, or query sources via the GraphQL API.

Where can I propose changes or report issues?

Use GitHub Discussions for ideas and questions, and the schema repository for schema issues.