Yahoo!: Web 3.0 is All About Desktop RIAs

Yahoo! Developer Network Technical Evangelist Mattt Thompson put up a long post today on the YDN blog laying out his vision for the future of the web and what Web 3.0 might look like. Says Thompson, Web 3.0 will be all about the ubiquity of desktop/offline access for rich Internet applications.

Thompson framed his case for Web 3.0 by defining Web 2.0 — or at least the technology that made it possible. According to Thompson, regardless of the specifics of the Web 2.0 definition that you subscribe to, most of what we think of as the current generation of the web was made possible by Ajax.

“Asynchronous communication between the browser and servers (a la Ajax) provided the latency needed to recreate the look and feel of a regular desktop application,” he writes. “Not to argue the finer points in this minefield of buzzwords and strong opinions, but the emergence of rich user interaction changed everything. It got people to rethink what a website could be. It was a newer, shinier series of tubes.”

Thompson thinks that other new technologies could have a similarly substantial impact on the web and “make Web 2.0 no more than a distant memory.”

His leading candidates right now are Gears — the open source offline sync API from Google — and Yahoo! BrowserPlus, a similar offline API for web developers that takes a different, more modular approach than Google’s Gears.

Thompson says that even though many bloggers and developers have described Gears and BrowserPlus as competitors, they’re not necessarily so. “With their respective design goals in mind, it doesn’t make sense to look at Gears and BrowserPlus as competitors. Sure, both have an intersecting feature set, including desktop integration, and automatic software updates, but that’s where the similarities end,” he says.

Gears is a back end technology, says Thompson, while BrowserPlus has potential to be used for front end tasks. Together, they reveal his vision for Web 3.0: an Internet in which applications move seamlessly from between online and offline environments, from the web to the desktop, and where that is as expected as all the Ajaxy-bits are today.

“If you think back to how Ajax changed the way people could interact with websites, much of it is completely transparent. For instance, deleting items in a list with real-time feedback is so pervasive that it became an expected behavior,” writes Thompson, who says that he can envision a web where the type of things enabled by Gears and BrowserPlus become “equally pervasive, yet [remain] invisible to the normal user.”

If RIAs are the future of the web — or even the future of computing — then offline/desktop access is supremely important. The browser is not a very good operating system, so being able to run applications on the desktop is an important transitional step toward Thompson’s vision of Web 3.0 and the future of thin client computing.

Free book: Jump Start HTML5 Basics

Grab a free copy of one our latest ebooks! Packed with hints and tips on HTML5's most powerful new features.

  • rickmans

    Well a vendor lock-in, that a new approach. Finally we were finished with the heritage of IE’s and now we get plugins that do not differ that much. Are there any open alternatives for these two plugins?

  • Mattt Thompson

    @rickmans As I wrote in my article, Gears was released as open-source in May. Unlike the browser wars of the 90′s, Gears is positioning itself as an extension of existing web standards, ie HTML5. The modular approach of BrowserPlus, however, does not easily lend itself to a completely open-source implementation as it necessitates a gatekeeper of sorts. However, Yahoo! and the BrowserPlus team are doing their best to open up the process as much as they can, without compromising security to users.

  • http://www.foundrylabs.co.uk stuartk

    @Matt, just wondering how you think Adobe Air fits into this picture. Although it’s not Open, it does allow a transition from web 2.0 to RIA’s, as I believe you can use html, flash or flex to develop with Air.

  • Mattt Thompson

    @stuartk Right now, I have a strong, but loosely held opinion about AIR. It seems to me that AIR is approaching desktop integration from the wrong direction. It falls somewhere between OS widgets / gadgets and real web applications, with none of the advantages. I just don’t see users being motivated enough to go through such an extensive installation process to try out something not much different than a widget or a normal website.

    What I’ve found to be much more interesting is the approach that Mozilla Prism and Fluid are doing by leveraging existing rendering engines to easily port heavily-used applications to specific applications.

  • http://impl.emented.com jobol

    Browser extensions can play an important niche role in pushing web standards forward, but are they important enough to deserve a Web 3.0 label? I’m not so sure. I wrote this post about a few other recent Web 3.0 proposals.