Chris Alexander

On Engineering

Where The Web Goes From Here

27th February, 2010

There is always talk of where the web could go next.

Constantly seeking the next “big thing”, entrepreneurs and large companies alike are pouring time, money, and intelligence into figuring out the next cool thing that can be done with the network that has the ability to connect us all.

But where can the web go from here?

The best answer to that question can be taken from extrapolating where we’ve come from into the future. Here’s what I think will be big in the coming months and years.

Web 1.0: The Corporations

And so, when the Internet was born; growing out of military research projects and originally connecting military and civilian research centers, it was a wild and open utopia of free access and shared knowledge.

Along came the internet behemoths; CompuServe, Prodigy, (discreetly shifting fowards to…) Yahoo! and AOL. They fed content down to the lowly internet user, defenceless against the unmanageable stream content pushed into even more unbearable looking pages.

Web content was a very one-way flow, with the general user being unable to fight this torrential content river; this marked Web 1.0.

(No alt specified)

Finally, users were given a way to filter out the ocean of content, spam, adverts and trash into a meaningful piece of data that could be used.

From the very early stages, search was always going to be massive. Did Larry and Sergey know what they were getting themselves in to?

(No alt specified)

Web 2.0: User Content Reigns

As we find ourselves riding the crest of this real-time surge, we have come a long way from the corporate rule of the early internet.

With the birth of blogging, users were finally given a tool to contribute content to the internet.

Then came Wikis, photo blogging, video blogging, life blogging, micro blogging and social networks, and suddenly there was more content than anyone could ever know what to do with.

(No alt specified)

Web 2.1: The Mobile Web

What’s the point in using the internet for socialising when you can only do it with your desktop computer, from home?

With the advances in mobile phones, 2G and 3G data networks, and innovators finally realising the mobile potential, the mobile web was a long time coming, but when it came it hit hard.

This isn’t just down to the iPhone of course; there was a thriving (although pretty expensive) mobile internet uptake before that. Now with mobile internet expected as standard on tariffs, and consumers expecting the phones to be able to handle it too, it is foolish of large websites not to provide a mobile version optimised for devices on the go.

(No alt specified)

Web 2.2: Realtime

The way it used to work was you wrote a blog post, and then eventually people found it and read it.

With content made available in near real-time on Facebook, Twitter, and most notably FriendFeed (the true innovators here, let’s not forget), it brought a whole new dimension to the social web.

(No alt specified)

Where Next?

That was a very quick and very general overview of some of the major milestones in getting here.

I was going to jump into Web 3.0, but there is just so much potential in 2.x still to come, I don’t think we could without taking a look first.

Web 2.3: Location

As I said in a tweet replying to someone who suggested location was the next big thing, I said it was the next big step: “We’ve done ‘me’ and ‘now’, next up is ‘here’.”

Twitter have geolocation built-in to every tweet; Foursquare and Gowalla (two social location-based games) are growing monstrously quickly, and Google have been offering localised searches for years.

Combine the real-time developments of the past couple of years with this targeting to where the user is right then, and you have a potent mix for providing some superbly targeted content (and advertising).

(No alt specified)

Web 3.0: Data and Nodes

So after we have all gotten very very creative with this data network, I think Web 3.0 will be the advent of an entirely new way of using the internet; let me explain.

Today, the network is structured like the information flow from the Web 1.0 days; ISPs provide data connectivity to the users below them.

Imagine what it would be like if the lower level nodes (mobile phones, laptops, desktop computers) were connected in the same peer-to-peer manner as the top level networks that currently span the globe; and how this would enable even smaller connected devices to very easily connect to the network.

Consider also the possibility of shifting from the website-driven services of today to a data-powered internet. For example, each node has its pre-defined set of data and permissions; each other node then can accessed depending on whether the other node has access rights. Consider this as an extension of the growth of APIs and data services over the past few years. The prime example of this is the Twitter API, which has seen so much growth that its format is being used by other services as a template for their data formats and endpoints.

Here’s a couple of sample applications to clarify:

The Road Network: Bringing a new meaning to “road network”, imagine if all the cars on a motorway were networked together through a peer-to-peer configuration. Bandwidth would be spread across the locally available nodes, and alerts for accidents would propagate instantly, rather than the hours it can take via LED signs and radio stations.

(No alt specified)

Cheap, distributed nodes: Imagine if each traffic light had a small node attached to it with a couple of sensors and a mid-range data connection. The nodes would be able to report information when queried, as well as form a peer to peer network to enable communication across the network.

(No alt specified)

What’s the point though? Aren’t we doing this already?

To an extent, yes.

However, if we were to increase the use of the innovative peer-to-peer architecture to the lower networking levels, we can increase network availability and reliability to distributed devices.

And the uniform data model, while people have tried before (SOAP, anyone? Eurgh), going for something at a lower level; more like you would find in the engine management system in your car (CAN, for example) - robust, reliable, built in querying and easily extensible.

What do you think?

This has just been me riffing on some crazy ideas I’ve had festering in the depths of my cerebrum for the past few weeks.

What do you reckon? Is location the next big thing? Will Web 3.0 be a series of nodes connected by peer-to-peer powered networks?