On the semantic web
A personal reflection on the semantic web, examining its original vision of machine-readable linked data and the challenges that prevented widespread adoption, including complexity and lack of incentives.
Background
- The **Semantic Web** is a vision (popularized by Tim Berners-Lee in the late 1990s) where web data is structured and labeled with explicit meaning so machines can understand, connect, and reason with it automatically, rather than just displaying it for humans.
- Key technologies include **RDF** (a graph-based data model), **SPARQL** (a query language for that data), **OWL** (ontologies for defining relationships), and **linked data** principles (using URIs to connect related datasets).
- Despite serious investment from W3C, academia, and companies like Google (Knowledge Graph, Schema.org), the Semantic Web never achieved mainstream adoption — critics cite its complexity, the burden it places on content creators, and the chicken-and-egg problem of needing rich metadata before useful applications can exist.
- More recently, **large language models (LLMs)** like GPT have revived interest in machine understanding of text, but they do it probabilistically rather than through formal logic and ontologies — raising the question of whether the Semantic Web's original approach is still needed or viable.