schema.org

schema.org is a shared vocabulary for describing things on the Web — people, organisations, events, places, creative works and many more — so that machines can read what a page is about, not only what it says. It was launched in 2011 by Google, Microsoft (Bing) and Yahoo!, soon joined by Yandex, and is maintained as an open community project at schema.org, hosted by a W3C community group.

How it works

The vocabulary is a hierarchy of types — Thing > CreativeWork > Article > ScholarlyArticle, for instance — each with its properties, whose values can be text, numbers, dates or other typed objects. A page describes itself in one of three syntaxes: Microdata and RDFa, which annotate the visible HTML, or JSON-LD, a separate block of JSON in a <script> element, which is the one search engines recommend:

Unlike Open Graph, which flattens a page into a few strings for a preview card, schema.org describes a graph: an article is part of an issue, which is part of a volume of a periodical; its authors are persons, who may be affiliated with organisations. The @id and sameAs properties identify the same thing across sites — a DOI, an ORCID, an OpenAlex identifier.

Types for research

type	for
ScholarlyArticle	a journal or conference paper
Chapter	a chapter in a book
Book	a monograph, an edited volume, a proceedings volume
Periodical, PublicationVolume, PublicationIssue	a journal, its volumes and its issues
Thesis	a thesis or dissertation
Report	a technical report
Dataset, SoftwareSourceCode	research data and software
Person, Organization	authors, editors, affiliations, publishers
Event, Course, Grant	conferences, courses, research funding

Who reads it

Search engines — Google, Bing, Yandex — use it to understand pages and, for some types, to show rich results. Google Dataset Search is built on Dataset markup.
Knowledge graphs and aggregators harvest it to link the same entity across sites.
Google Scholar relies on the citation_* tags instead. Social networks read Open Graph. The vocabularies serve different readers and are used side by side.

A page can be checked with the Schema Markup Validator, which tests the vocabulary, and Google's Rich Results Test, which tests what Google makes of it.