The Management of the Web Platform
As running software, the Web is represented by three browser engines: Webkit, Chromium, and Gecko. Each of these is owned by a corporation — Apple, Google, and Mozilla — but licensed as open-source software and developed in the open: subject to the details of the chosen license, anyone can use, and contribute to, the code. But, it is their basis in shared standards — specifying the features shared across projects — that makes these three separate software projects represent the Web. Where did these standards come from?
Once upon a time, somewhere early along the timeline of the Web’s history, two competing factions ruled the land of the Web. One faction, consisting of a majority of members of the World Wide Web Consortium (W3C), wanted to leave HTML behind and replace it with a new, promising, but untested technology. The other faction, consisting of the implementers of early browser engines — Apple, Mozilla, and Opera — wanted instead to stick to the tried and tested HTML, and evolve it as necessary. A schism followed, and the implementers created a venue named the Web Hypertext Application Technology Working Group (WHATWG). Not breaking the Web(by evolving HTML rather than replacing it), ensuring that specifications reflected reality(even if it meant changing them), and a high level of specification detail(allowing engines to match each other’s behavior)— these were he basic principles upon which the WHATWG was founded. The implementers — later joined by a new entrant to the web engine market: Google — then started working on a new version of HTML, which initially came to be known as the buzzword HTML5, and later, when it dawned upon people that version numbers were not helpful as progress was gradual and continuous, simply as HTML. It was only 15 years after the schism that the WHATWG and the W3C agreed to work together on the WHATWG’s version.
Today, the Web is ruled by a spirit of collaboration, and standards are published by both organizations. The HTML standard remains the mother of all standards, but, it finds itself supplemented by a host of smaller ones: each dedicated to one modular feature of the Web and supported by a working group. Experimental features go through an incubation stage, and, if enough support is garnered, move on to standardization. The entire universe of standards is backed by a suite of automated tests, which each implementer uses to ascertain the status of its engine on the continuum of implementing the Web. A constant interaction occurs between standards, implementations, and automated tests: sometimes an implementation is changed to reflect a standard, sometimes a standard is changed to reflect an implementation(example). On both sides, the process is never considered finished, and numbering versions of the Web has long been abandoned.
It is now time to delve deeper in what a Web engine exactly is, which we will do next.
