Deutsch
Basics & formats

Semantic Web

Also known as: Web of Data

The Semantic Web is the W3C’s concept of a web in which data is described in a way machines can process, so it can be shared and reused across applications, companies, and communities. For SEO, the idea lives on mainly in structured data based on schema.org.

Where does the idea come from?

Tim Berners-Lee outlined it in his “Semantic Web Road map” in 1998, describing a web of data that works in some ways like a global database. In 2006 he added four Linked Data principles: use URIs as names for things, use HTTP URIs so people can look them up, return useful information in standard formats when someone does, and link to other URIs.

Which standards belong to the Semantic Web?

The foundation is RDF, which stores statements as triples of subject, predicate, and object. RDF Schema and the ontology language OWL define classes and relationships on top of it. SPARQL is the query language for RDF data. The upper layers of the original “layer cake” diagram, such as proof and trust, were never standardized. In December 2013, the W3C folded its Semantic Web Activity into a broader Data Activity.

Entities and relationships in a knowledge graphFictional example with four entities: the organization Example Inc., the person Jane Doe, the place Sampletown and the product Example App. Edges connect them with schema.org properties: founder leads from the organization to the person, location to the place, manufacturer from the product to the organization. Through sameAs the organization points to its Wikidata entry and its Wikipedia article. On the right is a knowledge panel that Google generates automatically from data like this.foundermanufacturerlocationsameAsWikidata entryWikipedia articlePersonJane DoeProductExample AppOrganizationExample Inc.PlaceSampletowngenerated automaticallyKnowledge panelExample Inc.CompanyHeadquarters: SampletownFounder: Jane DoeProduct: Example AppEach edge is a statement of subject, predicate and object (an RDF triple),e.g. Example Inc. · location · Sampletown. All names are fictional.
A knowledge graph describes entities such as people, places, organizations and things and connects them through named relationships. sameAs points to pages that unambiguously identify an entity. Google generates knowledge panels automatically. Fictional example; how Google stores entities internally is not public. Sources: Google for Developers: Knowledge Graph Search API; schema.org: sameAs; Google Help: About knowledge panels; W3C: RDF 1.1 Concepts and Abstract Syntax (triples).

What does the Semantic Web mean for SEO?

The full vision of machines reasoning across the web did not materialize (assessment). Parts of it are in daily use. Schema.org says its data model is derived from RDF Schema. Google introduced its Knowledge Graph in 2012 with the phrase “things, not strings,” focusing on entities instead of keywords. Marking up an organization or product with JSON-LD describes an entity and its relationships in that spirit.

Example

JSON
{
  "@context": "https://schema.org",
  "@type": "Organization",
  "name": "Example Inc.",
  "sameAs": "https://www.wikidata.org/wiki/Q0000000"
}

The sameAs property links the organization to an external description of the same entity.

Sources

  1. W3C Semantic Web Activity (archived page, frozen December 11, 2013)w3.org
  2. Semantic Web Road map – Tim Berners-Lee, W3C Design Issues, September 1998w3.org
  3. Linked Data – Tim Berners-Lee, W3C Design Issues, 2006w3.org
  4. RDF – Semantic Web Standards, W3Cw3.org
  5. Data model – schema.orgschema.org
  6. Introducing the Knowledge Graph: things, not strings – Google, May 16, 2012blog.google

Last updated: