In all fictional accounts of the metaverse, the immersive 3D world that constitutes the user interface to virtual reality overcomes some fundamental limitation of a more traditional online experience, and the barriers to entry are minimized away. In Neil Stephenson’s Snow Crash, or any of William Gibson’s novels, for example, the net is an information-rich environment, and the “presence” of the stories’ protagonists in virtual space is a way of providing a higher bandwidth to transfer that information to the characters. The problems of a user interface are gone, vanished into some magical direct-brain interface that overcomes the limitations of screen resolution and archaic keyboards. Content ownership issues and ownership rights are nowhere to be seen. When those issues are even hinted at, the metaverse appears to run according to classic libertarian theory: rights exist where they are defensible and defended; reuse is rampant; the economics of baseline development are ignored (Who paid Hiro to write the sword fighting program for The Black Sun? How did he eat while he did it?)

Compared to this, SecondLife looks pitiful. The interface is a huge barrier to entry; a substantial amount of time is required just to learn basic movement. (In fairness, Stephenson hints at a similar learning curve in Snow Crash, but only in passing.) Linden Lab attempted to improve it with viewer 2.0, a well-intentioned but impossibly misguided effort which miraculously, simultaneously fails to improve the experience of seasoned users, and at the same time manages to raise the bar for new residents even higher. The result was horrific: a 38% reduction of staff at Linden Lab, and the head of M Linden, a victim perhaps of his own failure to deliver.

Viewer 2.0 went wrong because it was based on a fundamental misunderstanding of what’s needed to take SecondLife beyond its current hobbyist user base. Linden Lab eyed the vast user communities of web 2.0 social networking sites as an unexploited market, and attempted to build a viewer that would trick all those Farmville and Mafia Wars players into signing up – and eventually paying – for a cool new(-ish) online experience.

Much has been written about how Snow Crash influenced Philip Linden’s vision during the creation of SecondLife. If the Lindens had kept going back to that novel, they might have caught on to something that seems obvious to me: what’s needed is not ways to get new users into SecondLife; instead, we need ways to integrate other net content into the SecondLife experience, creating a compelling experience that will attract users to the platform. SecondLife should provide a means for its users to access and interact with other parts of the internet in-world, in ways that, if they are not easier than RL, are at least somehow more compelling. In practical terms, as a developer and content producer, this means: SecondLife needs a stable development environment that integrates with a variety of standard development tools, and it needs APIs and service points that allow developers to pull all manner of content into the in-world experience. Rather than trying to get Facebook users to join SecondLife, how about

• integrating Facebook (and other) profiles with SecondLife avatar profiles
• providing reliable interfaces to external calendaring services
• providing similar interfaces to Twitter, Blogger, and other forms of content
• abandoning the in-world search, and replacing it with Google (by making all in-world content discoverable via the web)
• replacing the HUD concept with a supported viewer plugin that allows third-party tools to modify the interface

Some of these things already exist in the form of scripted objects that are available in SecondLife. Viewer 2.0 has support for media plug-ins, but I’ve not seen anything that integrates with the viewer UI. None of them have the level of support, reliability, or client integration that is needed to make the experience compelling for a new user. If we’re going to use SecondLife to browse blogs, read RSS feeds, post twitter updates, or read MySpace pages, there has to be something that differentiates that experience from firing up a web browser for those kinds of things.

To find that, though, we will have to answer the fundamental question that lurks behind all discussions of SecondLife: What is SecondLife for? That we even have to pose this question should raise a red flag immediately. There have been a lot of talk about SecondLife as a platform for business applications, but let’s get real: there are exactly two ways to potentially make money in SecondLife:

• Rent object storage and virtual real estate from Linden Lab, and then sublet that out to other users, or
• Build compelling content (either permanent builds or events) and charge other users for access to that content.

In either case, the end-consumer of SecondLife is… the current user base. There’s no “pull” to bring people in from the outside, other than the “cool factor” of SecondLife (and its user-generated content.) Linden Lab has been searching for a way to make SecondLife a marketable product, and failing miserably. For example:

• As place for training and online meetings, SecondLife fails: Virtually any web-based solution is easier to use both for presenters and participants.
• As a place to market real-world products, SecondLife fails: The web reaches a huge segment of the world, in one form or another. All the grids and platforms combined with SecondLife are a tiny speck by comparison.
• As a venue for ‘live’ performance, SecondLife fails: even if watching Bono’s avatar was as cool as seeing streaming video of Bono, the fact that we can’t get more than seventy people onto a sim without it crashing is literally a show-stopper.

SecondLife remains a niche, and it will continue to do so. It serves specific communities well, but only because the members of those communities are willing to overcome the barriers to entry and overlook the limitations of the platform.