Development of HTML stopped in 1999 with HTML 4. The W3C focused its efforts on changing the underlying syntax of HTML from Standard Generalized Markup Language (SGML) to XML, as well as completely new markup languages like Scalable Vector Graphics (SVG), XForms, and MathML. Browser vendors focused on browser features like tabs and RSS readers. Web designers started learning CSS and the JavaScript™ language to build their own applications on top of the existing frameworks using Asynchronous JavaScript + XML (Ajax). But HTML itself grew hardly at all in the next eight years.
Recently, the beast came back to life. Three major browser vendors—Apple, Opera, and the Mozilla Foundation—came together as the Web Hypertext Application Technology Working Group (WhatWG) to develop an updated and upgraded version of classic HTML. More recently, the W3C took note of these developments and started its own next-generation HTML effort with many of the same members. Eventually, the two efforts will likely be merged. Although many details remain to be argued over, the outlines of the next version of HTML are becoming clear.
This new version of HTML—usually called HTML 5, although it also goes under the name Web Applications 1.0—would be instantly recognizable to a Web designer frozen in ice in 1999 and thawed today. There are no namespaces or schemas. Elements don't have to be closed. Browsers are forgiving of errors. A p is still a p, and a table is still a table.
At the same time, this proverbial unfrozen caveman Web designer would encounter some new and confusing elements. Yes, old friends like div remain, but now HTML includes section, header, footer, and nav as well. em, code, and strong are still present, but so are meter, time, and m. img and embed continue to be used, but now there are video and audio too. However, closer inspection by the caveman designer would reveal that these elements aren't that different. Many of them might be things the designer needed back in 1999 but didn't have. All these new elements are easily learned by simple analogy with elements the designer already understands. In fact, they're a lot easier to learn than Ajax or CSS.
Finally, when the caveman fired up the 300MHz laptop running Windows 98 that was also frozen in 1999, they might be astonished to realize that the new pages display fine in Netscape 4 and Windows® Internet Explorer® 5. Sure, the browser wouldn't recognize or do anything with the new elements, but the page still displays, and the content is all there.
That's not a happy coincidence. HTML 5 was explicitly designed to degrade gracefully in browsers that don't support it. The reason is simple: We are all cave people. Browsers now have tabs, CSS, and XmlHttpRequest, but their HTML renderers are stuck in 1999. The Web can't move forward without accounting for the installed base. HTML 5 understands this. It offers real benefits to page authors today while promising even more to page readers tomorrow as browsers are slowly upgraded. With that in mind, let's look at what HTML 5 brings you.
Even well-formed HTML pages are harder to process than they should be because of the lack of structure. You have to figure out where the section breaks go by analyzing header levels. Sidebars, footers, headers, navigation menus, main content sections, and individual stories are marked up by the catch-all div element. HTML 5 adds new elements to specifically identify each of these common constructs:
section: A part or chapter in a book, a section in a chapter, or essentially anything that has its own heading in HTML 4header: The page header shown on the page; not the same as the head elementfooter: The page footer where the fine print goes; the signature in an e-mail messagenav: A collection of links to other pagesarticle: An independent entry in a blog, magazine, compendium, and so forthFor example, consider a typical blog front page with a header at the top, a footer at the bottom, several entries, a navigation section, and a sidebar, as shown in Listing 1.