About
Over its fifteen year history, web browser technology has passed through three distinct phases. The popularization of the World Wide Web that followed the release of NCSA Mosaic in 1993 was driven by two hitherto obscure innovations: markup-based documents and hypertext. Neither of these was really new, of course. John F. Kennedy had just announced his intention to run for President of the United States when Ted Nelson launched Project Xanadu, a hypertext document system that remained a dream for the better part of 40 years. The Beatles were breaking out as a hot new rock band when Charles Goldfarb and others began tinkering with document markup at IBM. Their work led eventually to SGML’s acceptance as an international standard in 1986. But it was with Mosaic and its successors that the idea of navigating a vast web of interlinked pages filtered into the general consciousness, transforming computing from a hobby for social misfits into a defining characteristic of modern culture.
In the beginning, the web was all about reading. The number of pages available was vast and growing at a stunning pace. The ability to jump with a single mouse click to a related document engendered a radically new way to consume information. But at its core, the web was still no more than a turbo-charged substitute for paper.
Around 1995, it entered into its second phase with the advent of the Common Gateway Interface. In the original 1993 specification of HTML, Tim Berners-Lee was already hypothesizing that his creation could be used to represent “database query results”. To do so, web servers needed a standard mechanism for calling out to an external application rather than returning a pre-existing web page. CGI filled this role, though it required that a new process be spawned for each request and evolved quickly into a number of more efficient approaches. The ability to generate an infinite variety of “dynamic” pages by querying a database based on criteria solicited from the user made possible a new generation of web sites, like Amazon and eBay, which resembled traditional software applications.
Server-based applications of this type are at the mercy of network latency. Even something as simple as sorting a list of names of the screen requires sending a further request to the server and downloading a whole new page. The boundary between desktop and web applications was crystal clear: the former were fast and responsive, while the latter gave access to vast volumes of data and could be accessed from any computer equipped with a web browser, but were frustratingly slow.
Two new innovations set out to remedy this problem. JavaScript was invented by Brendan Eich and released in late 1998 as part of Netscape Navigator. With JavaScript, software developers could add logic to their applications that ran on the client. Features like the above-mentioned sort could thus be implemented in a way that didn’t require a round trip to the server. XMLHttpRequest was added to Internet Explorer in 1999. Its original usage was in Outlook Web Access 2000, where its ability to request data from a server in the background, without loading a new web page, was put to great use.
JavaScript and XMLHttpRequest made possible a new class of web applications that retain the advantages of the previous generation while offering responsiveness more comparable to that of desktop apps. Instead of loading a new page every time the user does anything, they grab data from the server as needed and do most of the heavy lifting on the client. The combination of JavaScript and XMLHttpRequest is so compelling that a clever coinage has come to describe the duo: Ajax (Asynchronous JavaScript and XML).
This new phase is epitomized by Gmail, which was one of the first Ajax applications. (Who would have thought that web-based email clients would play such a crucial role in the evolution of web browser technologies?) It has been accompanied by a wave of web sites that aim first and foremost to help users interact and express themselves. The whole movement is often referred to as Web 2.0, although my nomenclature would make it Web 3.0. This disagreement is likely to be moot quite soon, in any case, since the future Web 4.0 (or 3.0, if you must) will probably be popularized under an entirely different moniker. Exactly what that might be is hard to say, since it’s not yet clear what the salient characteristics of the web’s next phase will be.
There is an increasing number of clues, however. For the first time in years, the web browser space is alive with activity. The aim of this blog is to draw an outline of the future web and then color it in piece by piece. I predict that 2008 is going to be a big year in this regard. If so, it’s going to be a lot of fun watching the saga unfold.
About Me
My name is Matthew Gertner. Currently I am a full-time contributor to Mozilla Prism. Until recently I was the co-founder and Chief Technology Officer of AllPeers. I like playing golf, backgammon and piano, cooking unreasonably ambitious meals, thinking and writing about technology, mastering foreign languages and the view of Prague castle on a crisp winter evening over the crest of the hill in the Riegrovy sady. And web browsers. Oh, and piña coladas.
Disclaimer
I’m an active member of the Mozilla community and big fan of all things Mozilla. I’m going to try not to be biased, but that might not always work. And that, my friends, is why blogs have comments.