Will the Web Browser Replace the Operating System Desktop?

Like most of geekdom, I sat transfixed in front of my computer yesterday, first following Engadget's live blog of Apple's WWDC keynote and then watching the video of the event. A large part of Apple's mind-bending success over the past few years can be attributed to the effort they have put into improving the user experience of the operating system desktop. Probably more even than the Mac's slick hardware, the superiority of OS X over competitors has been driving users into Apple's arms. Even more striking is the quantum leap that iOS (then iPhone OS) represented when it was launched in 2007. The iPhone has great hardware as well (the capacitive touch screen was particularly revolutionary), but iOS was clearly a major reason why so many people went bonkers over the new phone.

Apple is not standing still. They are continuing to improve the user experience of both of their operating systems. OS X is absorbing many of the most successful features of iOS while adding an array of other features like document versioning and better access to running apps. iOS has much better notifications, big improvements to Safari and a bunch of other stuff. You could argue that this is just so much unnecessary cruft designed to sell OS upgrades, but I don't think so. I'm looking forward to a lot of the new features on both platforms because they will make a lot of my computer usage faster, easier and less error-prone.

Today I saw the announcement of a Webian, a prototype based on Mozilla Lab's Chromeless initiative that aims to replace the operating system desktop with a browser-based interface. As the program's author, Ben Francis, explains: "If you’re anything like me then you’ll find that most of the stuff you do on your PC these days happens in a web browser, and the desktop environment you used to depend on is now just getting in your way."

There are certainly some merits to this idea. By getting rid of the OS desktop, we sever our ties with the physical computer we are working on. Everything can live in the cloud (are we still allowed to use that cliche?), and applications are downloaded and cached as needed, instead of cluttering up our disk until we eventually give up and buy a new computer. This means we can sit down at another machine, fire up our browser-based OS and experience exactly the same desktop environment as we do at home or in the office. Indeed, the notion of the browser becoming the OS has been around almost as long as the web, with Larry Ellison hawking his Network Computer as early as 1996. Google is making a big bet on this vision as well with Chrome OS, as most of the commentary about Webian was quick to bring up (the browser wars are back, baby!).

And yet I guess I"m not all that much like Ben, since I can't bring myself to get excited about this idea. It's not so much that a browser-based OS would lack so many of the features of a modern operating system like OS X, Windows or Ubuntu. Less features is often a very good thing in software design. But would you sell your car and ride around town in your refrigerator? It's question of applying the right tool to the right task. Of course, the browser could be extended with all the OS desktop goodness, but why bother when there are billions in R&D being invested into ongoing OS development efforts by tech giants like Apple and Microsoft?

The truth is that the premise of the browser-as-OS gets it backwards. Applications built using web standards, downloaded on demand from the web, have a lot of advantages over traditional apps. And sure enough, the software world is shifting inexorably in this direction. What is needed is better integration of web apps into slick, modern operating system desktops, not the wholesale replacement of the latter with a souped up web browser.

Matt

Matt