I thought Neal Stephenson‘s Snow Crash vision of a highly immersive consumer virtual reality Metaverse would soon become a reality, and I wanted to be part of it.

As we all know, Snow Crash crashed in only a few years. In August 2007, as CEO of a Second Life development company in Spain with some high profile clients, I was interviewed by the national TV to comment on why companies were wasting millions on desert corporate VR spaces without visitors. The “desert Second Life” meme, started by an article on Wired, propagated very fast, and Second Life started to fade out.

These days I have the impression that Second Life (SL) is a dead alien world populated by the ghosts of a few former inhabitants who refuse to go. A couple of weeks ago I helped to organize a very good SL conference with many good speakers and interesting talks… and less than 10 people came.

Looking back, I see several reasons for the demise of SL:

1. The interface is far, far too difficult for today’s casual users who think that the Internet is that little box with lights that flicker when you are on Facebook. Many users, including many who use the Internet daily for work, do not know how to copy and paste, or the difference between left and right click. Add to this a terminal attention deficit, and you see how the Second Life interface is too difficult for mass adoption. Today, you need to design one-click user interfaces, because two clicks is too many.

2. Related to 1, SL is too heavy for most user computers. Those with powerful gaming systems and modern graphic cards never realize it, but Second Life is just not usable on low-performance computers, including new computers of users who don’t know how to switch off resource-hungry antivirus software, firewalls, background tasks and all the useless crap installed by manufacturers.

3. A 3D interface that imitates reality can be a great and intuitive user interface (if you see a door, you should go through, if you see a chair, you should sit down, etc.), but 3D on a flat 2D screen is not really 3D, and may make things difficult for the user, especially when combined with 1 and 2.

4. Many early users of SL were very jealous and protective of the early SL culture, strongly centered on pseudonymity and non-disclosure of real life information, and vocally resisted all technical innovations that could facilitate the intrusion of reality into their “magic circle” (see for example the very heated debates that followed the introduction of voice in SL in 2007). Most of them were “immersionists,” mainly interested in SL as “another world” where they could live “another life” entirely separated from their “first life” (FL) and strongly resisted the “invasion” of “augmentationists” interested in SL as a communication tool for telepresence applications related to FL. I think the tension between these two communities played a significant role in the demise of SL. Henrik Bennetsen’s essay on the subject is not available anymore at its original URL but a backup is still on the Internet Archive Wayback Machine.

5. In my (and many others’) opinion Linden Lab, the company behind SL, made one disastrous decision after another, alienating existing users without attracting new users. For example, first they alienated immersionists by promoting SL as a platform for business and education, then they changed their mind, then they changed their mind again, then they changed their mind some more times until everyone stopped caring.

I wonder whether SL and other immersive virtual worlds still have a future, and whether Stephenson’s Metaverse can still become a reality.

I am interested in online communication tools, and I have no interest in an building an alternative life in another world. My ideal avatar is a simple geometrical primitive with realtime webcam video streaming on a face (like the default avatars in OpenQwaq). This is easy to do in SL since version 2 of the viewer, and it is easy to set up videoconferencing in SL with free video streaming services, but these things don’t seem too interesting for the remaining SL users. Perhaps they will catch on in the future in SL or (more likely) in next generation platforms.

There are, in fact, interesting next generation Metaverse platforms that run natively in modern browsers without requiring a dedicated viewer or plugin. See for example Cloud Party, or the awesome Virtual World Framework demos. These systems seem much lighter and easier to use than SL, and (hopefully) able to address points 1 and 2 above. Concerning point 3, consumer VR glasses like Vuzix Wrap, Oculus Rift and future version of Google Glasses will permit real 3D interfaces to 3D scenes, and may wake up the sleeping Metaverse. In summary, I am cautiously hopeful in a Metaverse renaissance.