There seem to be a lot of twos in the industry at the moment. From a personal perspective, I am coming up to the 20-year anniversary of having completed my Computational Science degree (with a gentlemanly 2-2, I might add), which means I have spent 2 decades in and around IT. Out there I hear lots of talk about Web 2.0, which is just going to be so much better than the cranky old Web it leaves behind (no, I don't believe this either), and meanwhile I look with fascination at the new generation of virtual collaboration environments exemplified by the all-encompassing Second Life, which will undoubtedly be so much better than the first (no, I don't believe this either).

Even between these two new online environments there is an oxymoronic duality. Let's consider them in turn: many initial versions of the "2.0" social networking and web-based integration tools have been clunky to say the least. Consider blogging: initial platforms were very much text-only, to be installed as a bunch of PHP scripts which required certain geeky configuration tweaks to work. This is certainly not a weakness; more, it is a symptom of provenance, in that a highly useful and yet ill-designed capability of many web sites—the log file—was then repurposed to achieve a level of previously unheard of interaction. On the integration side (indeed, we could define the 2.0 phenomenon in terms of interaction and integration), initial mashups have either been by select groups of individuals that have the time and/or the inclination, or monolithic single-use facilities that everyone has used ("oh look, another Google Map." Yawn). In other words, need/desire has preceded function, which is running to catch up.

I'm certainly not complaining—and furthermore, we've seen this approach before. Remember the first incarnations of email, totally based on text? Remember the first word processors or the Simple Messaging Service (SMS)? Remember bulletin boards, USENET or the various IP-based information sharing mechanisms that preceded the use of HTML? The advantage of function following form is that it can evolve to fit the usage models that develop. The adoption rate is faster, the result is (generally) slicker, and generally, the effects of proprietary lock-in and standards in-fighting are minimised.

Another approach entirely is to push technology to its limits, and then hang around waiting for people to work out what to do with it. We've seen this with broadband (particularly of the mobile kind), and now we're seeing it with immersive environments such as World of Warcraft or Second Life. The former is a computer game, and therefore has some inherent purpose; with the latter, purpose is a construct that needs to be created, like everything else in the 3D "virtual" environment. I know some people that have taken a look at Second Life and said, "what's wrong with the first life?" even while others think it's the best thing ever—the difference between the two schools of thought seems to be an appreciation of usefulness, or otherwise.

Just as form follows function when technologies are "pulled" by their users, so form precedes function when technologies are "pushed" by the industry. Neither approach is wrong: indeed they are just different sides of the same coin. Where things get very interesting however, is where the two sides converge.

Let's consider one specific limitation of virtual environments, be they gaming worlds or otherwise. When I first experienced World of Warcraft I discovered that there was a social element attached; the in-game equivalent of multi-channel instant messaging is rife, between both individuals and groups. Having spent more time in the game I realised that for many, WoW was a social networking environment that happened to have a game attached. I might not play as much as I used to but I still venture in occasionally, as much to find out how people are as to slay any dragons. I would imagine—though I've never spent long enough to find out—the same could be applied to Second Life.

What lacks from all these virtual social networks is any cross-linkage. There can be no following of a meme from Guild Wars to WoW, for example, or transfer of rights from Eve to Second Life. The demand is there: an April Fool that described an interoperability standard between the environments received a massive number of hits. However, the reality is largely that what happens in the game, stays in the game. Similarly, the limited APIs currently allow only the most limited of cross-platform integrations. There are mechanisms to output feeds to sites such as Thottbot for example, but the traffic can only ever be one way. Many in-game groups have their own Web sites but, at the moment, never the twain shall meet.

It doesn't take a rocket scientist to work out just how much more pervasive virtual worlds might be if integrated with social networking tools such as Twitter or MSN Messenger. Conversely, of course, there are undoubtedly some advances to be made in terms of bringing some of the capabilities of such tools into the Web 2.0 realm. There are plenty of companies exploring the use of 3D as part of how they serve up information to their customers: is it really such a leap of faith to imagine choosing an aeroplane seat by being presented with a Second Life representation of a plane, that links one's avatar to one's booking reference? Or indeed, taking a virtual tour of a new build development, or just being shown directions? Meanwhile, just when will it be that a rate of exchange is agreed between environments, such that virtual gold can be passed from one virtual world to another?

The Internet is all about breaking down boundaries, so it is ironic that one of the most rapidly growing usage scenarios should be limiting its own potential by imposing artificial restrictions. One thing's for certain however—such constraints never stay around for long. Its only a matter of time before the virtual doors come crashing down.