MF2

MF2, short for Microformats2, is a web standard for embedding metadata in HTML using class attributes and, to a lesser extent, rel attributes. It aims to make common kinds of data on web pages—such as people, articles, events, and organizations—both human-readable and machine-extractable without requiring separate data formats.

The core idea of MF2 is to annotate content with type and property classes. Types are indicated by classes such as h-entry (a blog post or article), h-card (a person or organization), and h-event (an event). Within a type, properties use prefixes like p-, u-, dt-, and e- to describe data: p-name (textual name), p-summary (short description), u-url or u-photo (URLs or images), dt-published or dt-start (date/time), and e-content (HTML content). The markup is plain HTML, and software can parse these class names to extract structured data.

MF2 was developed by the Microformats community as an evolution of earlier microformats, with notable involvement from Tantek Çelik. It emphasizes simplicity, backward compatibility with existing HTML, and ease of authoring. In practice, MF2-marked pages can be read by human visitors and by parsers that look for specific h- types and p-, u-, dt-, e- properties. It has found use in personal and community web projects, particularly within the IndieWeb ecosystem, where it also supports identity conventions using rel="me" links across sites.

While MF2 competes with other semantic formats like JSON-LD and RDFa, it remains a lightweight, human-friendly approach favored for straightforward HTML markup and data portability within microformats-friendly ecosystems. See also Microformats, Microformats2, IndieWeb.