contributed articles

DoI: 10.1145/1461928.1461945

2^W is a result of the exponentially growing
Web building on itself to move from a Web
of content to a Web of applications.
BY t.V. Raman
toward 2,
w
Beyond
Web 2.0
fRoM itS incePtion
as a global hypertext system,

the Web has evolved into a universal platform for deploying loosely coupled distributed applications. As we move toward the next-generation Web platform, the bulk of user data and applications will reside in the network cloud. Ubiquitous access results from interaction delivered as Web pages augmented by JavaScript to create highly reactive user interfaces. This point in the evolution of the Web is often called Web 2.0. In predicting what comes after Web 2.0—what I call 2^W, a Web that encompasses all Web-addressable information—I go back to the architectural foundations of the Web, analyze the move to Web 2.0, and look forward to what might follow.

For most users of the Internet, the Web is epitomized by the browser, the program they use to log on to the Web. However, in its essence, the Web, which is both a lot more and a lot less than the browser, is built on three components:

URL. A universal means for identifying and addressing content^{6, 7};

HTTP. A protocol for client-server communication⁵; and

HTML. A simple markup language for communicating hypertext content. ⁸

Together, they constitute the global hypertext system. This decentralized architecture³⁵ was designed from the outset to create an environment where content producers and consumers come together without everyone having to use the same server and client. To participate in the Web revolution, one needed only to subscribe to the basic architecture of Web content delivered via HTTP and addressable via URLs. This yielded the now well-understood network effect that continues to produce exponential growth in the amount of available Web content. In the 1990s, the browser, a universal lens for viewing the Web, came to occupy center stage as the Web’s primary interface. Deploying content to users on multiple platforms was suddenly a lot simpler; all one needed to enable universal access was to publish content to the Web. Note that this access was a direct consequence (by design) of the underlying Web contract, whereby Web publishers are isolated from the details of the client software used by their consumers. As Web browsers began to compete on features, this began to change in what became known as the browser wars, 1995–1999 ; browser vendors com-

36

peted by introducing custom tags into their particular flavors of HTML. This was perhaps the first of the many battles that would follow and is remembered today by most Web developers as the blink and marquee tag era marked by visual excess.

In 1997, HTML 3. 2 attempted to ease the life of Web developers by documenting the existing authoring practice of the time. HTML 3. 2 was in turn followed by HTML4²⁸ as a baseline markup language for the Web. At the same time, Cascading Style Sheets (CSS) ⁹ were introduced as a means of separating presentational information (style rules) from Web-page con-

 

52 CommunICatIons of the aCm | feBRuaRY 2009 | vol. 52 | No. 2