Death to Snow Crash

Like a lot of other people, I read Neil Stephenson’s novel Snow Crash back in the mid-1990s (along with other staples like Neuromancer and the still-applicable True Names). Like many, I was entranced by the idea of digital avatars with detailed facial expressions (something we were working on in 1995 and continue to today), and by the idea of  the ‘Metaverse’ – of having a digital home in a bustling virtual world that was somehow entirely immersive, that moved beyond visual and auditory to the kinesthetic. I was so taken with the idea as presented that I was willing to overlook its technical faults, and like so many others, dreamt of a huge all-inclusive world.

Many people continue to hold out hope for some Snow Crash analog as an all-encompassing virtual world: mostly this is discussed as the transfer of the Web to a VW, typically imagined in 3D, sometimes as a vision of something like Second Life (albeit more open, more interesting, with better performance, and maybe less of an emphasis on sex). This Snow Crash Metaverse is the online equivalent of the “flying cars” view of the today as seen from 1935. The latest installment in this vague, hand-waving exercise in techno-fantasy comes to us in the Business Week article, "Just Ahead: The Web As A Virtual World,” which enthusiastically describes the typical justifications for a 3D Web – buying jeans in 3D, walking from one web site to another, checking out a 3D virtual mall or hotel room, etc. – without more than a glance at the crippling issues and inefficiencies pipedreams like these present.

So it needs to be said: Death to Snow Crash. Death to the sugarplum visions of the 3D Web World that dance in our heads.  It’s time to move on.

Now I know Business Week is the mainstream media, writing for people who mostly don't spend their days up to their necks in virtual worlds.  I also know that there’s also a more informed group of people (some of them readers here) actively working on a ‘Metaverse Roadmap’ or toward a ‘3D Web,’ and that any criticism of this grand vision may seem like nothing more than curmudgeonly Ludditism. And it may well be that elements of 3D worlds will find their way onto the Web. But overall, I think the vision of the unified 3D-Web-Metaverse is one rooted in the past, not looking to the future.

If the ascendance of the Web and especially the changes brought by Web 2.0 sites, applications, and tools show us anything, they show us that the view of the Web as a singular thing is a mistaken continuation of an older centralized view. It also leads to the imposition of a set of inapplicable geographical constraints: information isn’t about place, and the Web doesn’t have a geography. We call this collection of online sites “the Web” for historical reasons, but it’s not really a web; it’s not even a unified thing like the virtual planet envisioned by Stephenson. Linking two html “pages” (another construct of convenience) does not create any form of geographical proximity. The Web may in fact be the least Euclidean, spatial, geographical construct ever made by humans.

Centralization isn’t the future; flexible decentralization is. Rather than trying to force-fit the Web into a Pangaea-like singular Metaverse (a huge sphere handed down from on high by the central authority of the ACM in Stephenson’s fictional world), we would do better to consider the future of many distributed worlds, some large and some small, some interactive 3D and some read-only 2D, and how they might combine into a mosaic of independent but potentially interrelated items, sites, and places. 

How do we get beyond the walled gardens of today’s virtual worlds? I don’t know. Yet. I’ve seen a few hints, but we have a long way to go. Maybe Vinge had it right with his archipelago of isolated but connected worlds in True Names; or maybe Gibson’s abstracted general cyberspace view giving way to internally more lavish worlds is more visionary and accurate. Or maybe – most likely – it’s something none of us have thought of yet.

But it’s not the “3D Web.” It’s not the Web as one big virtual world, or loading down typical web sites with the overhead of 3D spaces and expecting avatars to walk from one to another. The sooner we free ourselves from that fictional, limiting, inapplicable, 20^th Century concept (and stop promoting it in the mainstream media), the sooner we’ll have a good chance of inventing the future, rather than being stuck trying to re-invent a vision from the past.