Moving to a Web OS

John Markoff published an interesting article today in the New York Times about the shift in software and operating systems from the desktop to the Web, in which I am quoted. The article focuses on the rivalry and different styles between Microsoft and Apple’s next-generation projects that attempt to tie desktop operating systems and the Internet together more closely. I have been tracking this trend for a while now — a trend towards the evolution of what I call a "WebOS."

In my view the coming WebOS will not live only on the desktop, rather it will be a web service that lives "in the cloud." Desktops will become views into it, rather than the center of it. The desktop PC era is almost over. We are entering a new era of mobility and plurality — our digital lives will be spread across multiple devices, most of which will be mobile. We will require access to everything, no matter what device we are on.

When a user logs onto any device — be it a laptop or a mobile device — they will connect to their account in the WebOS. The local device will synch with their WebOS account to get their latest desktop layout, their preferences, and any new notifications or changes.

End-User access to the WebOS will be primarily through browser-based applications written in scripting languages, or running on server-side apps written in Java, C# or Ruby, rather than native desktop apps. Cases where native desktop code may still be needed will include high-end graphics and audio processing, or numerical calculations, that require a lot of computation. But for most consumers, such high-end needs are rare, except in the cases of gaming and multimedia. With the increase in mobile broadband and improvements in user-interface technologies, it will become less necessary to have native desktop code for such experiences — more and more of this will move to the Web. When native computation is needed it will take place via embedding and running scripts in the local browser to leverage local resources, rather than installing and running software locally on a permanent basis. Most applications will actually be hybrids, combining local and remote services in a seamless interface.

Once connected, the WebOS will provide users with a single point of access to their data, their relationships, their preferences, and their applications, anywhere, anytime, on any device. It will also begin to unify, or at least integrate, the data and functionality of different online and desktop applications in what will appear to the end-user to be "one place." Even though we may have accounts, data and relationships in many different services around the Web, our WebOS will provide us with a unified, centralized way to access this information. It will reduce the fragmentation in our digital lives and help to improve our productivity.

Imagine being able to go to one place on the Web to access all your email, documents, photos, videos, contacts and social relationships, RSS, data records, bookmarks, notes, and any other kind of knowledge or information. Imagine also that in this place you could also access all your "applications" — which themselves would be modular widgets or bits of functionality provided by various different web services and app developers around the Web. Imagine that in this place it would be easy to create new data types, populate them with data, and share them with others. Imagine that it would be just as easy to create new applications that could use that data, and share them too.

Think of the WebOS as the ultimate personal mashup. It would not matter anymore where information was actually stored — it could live in the cloud on the Net so it was available 24/7, and it could also be cached onto local devices like phones and laptops so that it was available locally or offline when needed. You could start to mix and mash your data in all sorts of new ways — you could for example see the connections between different kinds of things, or you could generate reports that might show for example, photos and videos by people you work with, or blog posts by your friends, or files related to meetings you are scheduled for, etc.

Because all information and application functionality would start to be integrated on a meta-level in the WebOS, new efficiencies in search, navigation and discovery would become possible. But to accomplish this there would need to be an easier and more flexible way to represent the data itself — a more open, extensible, remixable data model. Enter RDF, SPARQL, OWL and the Semantic Web. I believe these technologies provide a data framework that can help to accomplish this vision.

This vision of a WebOS is something I have been wishing for, and working towards, for a long time. My own startup, Radar Networks, is actually building something like this, based completely on RDF and the Semantic Web. Stay tuned! We plan to go beta in the fall. If you are interested, visit our website and sign up for our mailing list to be invited for early-access.

Social tagging: >

2 Responses to Moving to a Web OS

  1. acheyer says:

    Nice article, I agree with your vision 100%! We will soon be coming out with Sourcemix, a truly collaborative, sharing-based web app development community based, that will also feature the hybrid server/client model you describe.
    I don’t know if you saw the recent 1.0 release announcement from (opensource web-based OS), but it’s getting to be pretty impressive.
    – Adam

  2. SuperBliss says:

    Hey Nova,
    What a great vision for the future of the web. Having all that information in “one place” will offer big challenges for development of new and effective data storage / sorting methods. An article in Scientific American about a new and efficient approach to handling massive network traffic entitled,
    “Breaking Network Logjams… an approach called network coding…at its core is the strange notion that transmitting evidence about messages can be more useful than conveying the messages themselves.”
    When you consider the vision of WebOS you describe, and how it is essentially a giant network of information. The principles of network coding might very well be applied with great success.