Web 3.0 - Looking Into The Future

While Web 2.0 is all the rage today, I think it is time to think about Web 3.0. What would be the key features of Web 3.0.

Web 2.0 rests on two pillars:
1. bringing traditional desktop applications interactivity to web applications / services
2. greater user participation via wiki, blogs, digg, tagging etc.

User Interface

I think Web 3.0 will explore newer paradigms of interactivity like clickless user interface and predictive UI (ability to predict what the user will do next based on past interactions etc.).

Clickless UI - Clickless UI is one of the several possibilities. The current user interface concepts needs a massive overhaul. While we are accustomed to the standard web interface, it is not the most productive. Take for instance the browser. A very frustrating aspect of modern browsers is the inability to control it easily with keyboard. A simple enhancement would be to assign a number to each keys and then use the number to browse a link.

Predictive UI
User interfaces are generally static in the sense that for a given choice you are presented with a fixed UI. A predictive UI will observe the behavior of the user and similar users and provide (or hide) options individually. Windows XP start menu does something like this at a primitive level.

Smenatic Web
I think, though very much anticipated, semantic web will not emerge strongly even in Web 3.0 timeframe. A lot of progress and acceptance of Web is due to its unstructured nature. Artificially imposing a structure can only hinder its growth. It will become too complicated for common man (or woman).

Peer to peer browsing & communication

Traditionally web communication uses client-server model. While it suffices for many purposes it doesn’t allow to easily capture the dynamic nature of human communications, which is peer-to-peer, and hence news. A better system would rely on easy connectivity to peers to pass on locally relevant news. News is normally locality and time sensitive.
Peer-to-peer systems would also allow wider spread of internet to remotest locations. In such a setup, at its simplest form, each browser should also be a web server which accepts information (upload) and also passes it on to adjacent servers. JXTA seems like a good framework to facilitate this. I know this sounds too abstract but lets consider this. Suppose I am browsing a java architecture article somewhere, suddenly a news flashes on my browser telling me of an accident nearby or that there is a landslide / earthquake (take your pick) nearby or even that someone is in danger. On the other hand suppose you are on the road and your car breaks down. You simply grab the image will your mobile and send a short message for help. The cellphone is out of range of any tower but you reach via bluetooth to a nearby driver whose cellphone automatically relays it again. Soon, due to the type of message, it is caught by police scanners, which search for alert / accident type of messages, and passed on to nearby patrol cars.

Service is the platform

I think Web 3.0 will primarily be services based where services are supported by massive grid platforms. This will ensure the high availability of services and also scalability on demand with pay as you use model. This fits in well with the services model.

What are your predictions for Web 3.0?