History of HTML & HTML5
HTML wasn't designed in a corporate lab with a five-year roadmap. It was invented by one person to solve one problem, then grew — messily, competitively, and eventually collaboratively — into the standard that now runs billions of pages. Knowing this history explains a lot of HTML's quirks.
1991: Tim Berners-Lee and the first web page
In 1989, Tim Berners-Lee, a scientist at CERN, proposed a system for linking documents over a network so researchers could share information easily. By 1991 he had built the pieces needed to make it real: HTTP (a protocol for transferring documents), URLs (addresses for documents), and HTML (a way to format and link them).
His original HTML had roughly 18 tags, borrowed largely from an existing document format called SGMLguid. It was meant purely for sharing physics papers — nobody involved imagined it would become the foundation of a trillion-dollar industry.
The rapid growth years: HTML 2 through 4
Version | Year | What it added |
|---|---|---|
HTML 2.0 | 1995 | First version standardized by the IETF; formalized forms and basic tags already in common use |
HTML 3.2 | 1997 | Standardized by the newly formed W3C; added tables, applets, and text-flow around images |
HTML 4.01 | 1999 | Introduced strict/transitional/frameset doctypes, better forms, and stylesheet support via <code><link></code>, encouraging (but not enforcing) a move away from presentational markup |
This was also the era of the "browser wars" between Netscape and Microsoft, where each vendor added its own non-standard tags and behaviors to outdo the other — a major reason cross-browser compatibility was such a headache for years afterward.
The XHTML detour
In 2000, the W3C pushed the web toward XHTML — HTML rewritten to obey strict XML rules. Every tag had to be properly closed, lowercase, and correctly nested, or the whole page could fail to render at all. The idea was discipline and consistency, but in practice it made HTML unforgiving for a web full of hand-written, imperfect markup.
The W3C vs WHATWG split
By the mid-2000s, the W3C was steering toward XHTML 2.0 — a clean-slate, backward-incompatible redesign. Browser makers thought this was disconnected from how the web actually worked and how real pages were built. In 2004, engineers from Mozilla, Apple, and Opera formed a rival group: the WHATWG (Web Hypertext Application Technology Working Group), focused on evolving HTML incrementally based on what browsers actually needed.
The W3C continued down the XHTML 2.0 path for a while, then eventually abandoned it.
The WHATWG's pragmatic, backward-compatible specification gained real-world traction as browser vendors implemented it.
In 2007 the W3C officially adopted the WHATWG's work as the basis for the next version of HTML, ending the standards split.
HTML5 and the "living standard"
HTML5 became a World Wide Web Recommendation in 2014, though browsers had already been implementing pieces of it for years before that formal date. It brought native support for audio, video, and drawing (canvas) without third-party plugins like Flash, meaningful semantic elements (<header>, <nav>, <article>, <footer>), and a forgiving parsing algorithm defined precisely enough that all browsers handle broken markup the same way.
Today, the WHATWG maintains HTML as a "living standard" — there is no "HTML6" on the horizon. Instead, the specification is updated continuously as new features are agreed upon and implemented, similar to how modern JavaScript (ECMAScript) now ships yearly rather than in big, rare versions.
Timeline at a glance
1991 — Tim Berners-Lee publishes the first HTML tags at CERN.
1995 — HTML 2.0 standardized by the IETF.
1997 — HTML 3.2, then HTML 4.0, standardized by the W3C.
2000 — XHTML 1.0 pushes HTML toward strict XML syntax.
2004 — WHATWG founded in response to the stalled XHTML 2.0 direction.
2007 — W3C adopts WHATWG's work as the basis for HTML5.
2014 — HTML5 becomes an official W3C Recommendation.
Today — HTML is a continuously updated "living standard" maintained by the WHATWG.
Understanding this history explains why HTML is forgiving rather than strict like XML, why some conventions (lowercase tags, closing tags) persist out of habit rather than requirement, and why "HTML5" doesn't really have a fixed, final version number. Next, we'll zoom out to see how HTML fits into the bigger picture of how the web actually works.