HTMLMarkup

HTMLMarkup is a proposed or hypothetical markup layer designed to annotate HTML documents with machine-readable semantic information and metadata without changing their rendering in browsers. It is intended to improve accessibility, searchability, and tooling.

Core concepts include that HTMLMarkup sits atop HTML and defines a minimal set of semantic elements and attributes that describe structure, roles, and metadata, while remaining compatible with existing HTML. Authors can add semantic annotations via data attributes, such as data-semantic="heading", data-importance="high", and data-lang, enabling tools to extract structure and meaning.

Syntax and transformation involve tooling that maps HTMLMarkup annotations to standard HTML output for rendering, or to other formats. The approach emphasizes non-intrusive annotations, backward compatibility, and well-defined validation rules to ensure predictable processing across environments.

Use cases for HTMLMarkup include content catalogs, accessibility pipelines, CMS integrations, and documentation generation. Tools can leverage HTMLMarkup to generate navigation structures, outlines, or alternate representations of content, aiding both authors and users.

Status and standards: As of now, HTMLMarkup is not an official standard. It exists mainly in experimental communities and as a concept in discussions about semantic annotation of web content. Practice varies; when used, it relies on agreed-upon vocabularies and consistent tooling to map annotations to usable outputs.

See also: semantic HTML, ARIA, structured data, microdata, JSON-LD.