Where is the Web going? And what does Web 2.0 mean anyway?
Wed, Aug 2 2006

By Jonathan Oxer, IVT Technical Director

I'm finally back in Australia and recovering from jetlag after a very eventful trip to the US. The trip started with a few days in California to visit friends at Google and run a training session for some of their staff (which was recorded by their AV team and the video has now been published online at Anatomy Of A Debian Package in case you're interested in a very technical 1 hour lecture on how Debian software packages are created!) before I headed up to Portland/Oregon for the O'Reilly Open Source Convention, otherwise known as OSCON. OSCON was a blast - lots of interesting speakers, over 2600 attendees, and lots of opportunity to chat to many of the movers and shakers in the software industry. I did three talks at OSCON and a book signing for "Ubuntu Hacks", so it was a pretty hectic time.

Which brings me to the point of this topic: the future of the web. I've written and lectured in the past about how a lot of software is now being delivered online as web-based applications accessed through a browser rather than running directly on your computer. For example, the traditional way to manage a collection of digital photos is using a piece of software such as iPhoto running on your computer, with your photos sucked off your digital camera and stored locally in iPhoto for viewing. The "web application" way to do it is with an online service such as Flickr, where you upload your photos to the site and manage them remotely using your web browser instead of locally using iPhoto.

The obvious advantages of the second approach include removing your dependence on a particular computer to access your photo collection - all you need is a computer with web access to log in to your Flickr account and manage your photos. You don't need to install any software or worry about backups, because your computer does nothing but run a browser. In that respect web applications free you from using a specific computer and let you work from any computer you happen to have handy as long as you have an Internet connection.

But those sorts of benefits have been bandied about many times, and what OSCON really drove home to me was the paradigm shift from "web as publishing medium" to "web as collaboration medium", and how the most successful websites and web services now are ones that enable their users to participate rather than merely consume.

Let me explain.

The big catch-cry from the early days of the web was that it empowered anyone to be a publisher. Low technical barriers to entry and low costs allowed anyone to speak their mind to a global audience. The reality wasn't quite that rosy: "low technical barrier to entry" really meant "can write HTML, create graphics, and open an FTP connection to a server". For an IT person that might indeed be a low barrier to entry, but for the average person in the street it's "F-T-what?".

So the first decade of the web was really just an extension of the existing publishing paradigm. Sure, it was easier to create a website than to publish a book, but the information-flow was still mostly one-way: sites were "published", and users "consumed" them. It's a very top-down approach. "Web 1.0" was just publishing for the little guy.

But "Web 2.0" totally turns this around. Most people talk about Web 2.0 in terms of web-based software as I mentioned a moment ago, but it's actually a far more fundamental change than that. Web-based software provides the tools that have enabled the paradigm shift, but the shift itself is all about destroying the traditonal one-way information flow and replacing it with two-way interraction. Advertising spiels are replaced by conversations. Communication no longer needs to flow from a company with the resources to create a website out to multiple consumers: it ebbs and flows between individuals as they discuss as peers rather than merely consume.

The first stage of the shift came with the blogging explosion. Blogs allow non-technical Internet users to publish their thoughts and opinions online simply by typing some text into their web browser. It's still "publishing", but the barrier to entry is extraordinarily low: you can create an account with a blog provider and two minutes later be publishing your thoughts to the world with no technical knowledge at all. There are now an amazing number of blogs, with specialist blog search engines such as Technorati tracking in excess of 50 million so far. Users publish their thoughts, then other users respond using their own blogs or by appending comments to the original blog, and all of a sudden the net is full of multi-threaded conversations happening right out in public in real time. It's the ultimate democratisation of the Internet. No censorship, no corporate control, just whatever each individual feels like saying.

For example, after my talks at OSCON I discovered that several people who attended them had posted commentary online in their blogs: Jay Pipes and Terry's Worklog.

But even blogs were just the start of the paradigm shift. Blogs were the bomb that levelled the playing field and let anyone come along and play the game their own way, but it's only now that we're seeing how much further it can go.

The big changes now are based on a phenomenon called "mashups", where web-based services such as blogs, online databases, mapping systems, auction sites, online stores, and anything else you can think of are being creatively recombined by end users in ways the original creators never imagined. In additional to the usual interface you see with your web browser, many of the most popular websites provide special hooks called an "API", or "Application Programming Interface", that allows external software to interact with them directly behind the scenes. That means third party developers can write a little program that uses the content or services on the website as a component in some other creation.

That may not make much sense just yet, but there's a more detailed explanation online at Wikipedia: Mashup on Wikipedia.

The best way to understand it though is to look at a specific example.

Imagine you wanted to create a website that provided a guide to dog-friendly hotels. You would need a list of the hotels, of course, and you'd also want some neat way to display the information geographically so travellers could find one in the area they want to visit.

Doing it from scratch would be a huge amount of work. Doing it as a mashup, though, builds on the services provided by various other sites to create something totally new: Dog Friendly Hotels.

The Dog Friendly Hotels site combines the Google Maps interface with pictures of dogs stored on Flickr. By cross-referencing the pictures with the geographic location at which they were taken the hotel information can be superimposed on a searchable map.

Or maybe suck auction data from eBay and superimpose it on Google Maps to create a site that helps you find a used car for sale nearby: Dude Where's My Used Car?.

Or go crazy and combine Google AdWords, Google Maps, hostip.info, Microsoft Virtual Earth, NASA, NOAA Weather Service, WeatherBug, Yahoo Geocoding, Yahoo Maps, and Yahoo Traffic to create a site that automatically determines the visitor's location and gives access to local weather information and weather cams: Weather Bonk.

These are just a few examples of the mashups that have been created recently. There's a huge list of them online at: Programmable Web.

Some are incredibly clever while others are a waste of time, but they are all indicative of a trend toward end users having more and more control over what they use online. In fact the term "user" isn't even correct anymore. We're not users: we're participants. This is our Internet, and the tools we use in future are likely to be created by us or our peers rather than faceless corporations with big pockets who want to decide what we see and do.

Let your imagination run wild. You're an artist and the Internet is your canvas.

Cheers :-)

Jonathan Oxer
Technical Director
Internet Vision Technologies