I answered this question on Quora recently, which turned out to be more of an epic answer than I intended. So here are the reasons that I think the Metaverse doens’t exist in the pure ‘Stephenson’ model:
The processing power hasn’t been capable until now
It requires HUGE processing power to have millions, even thousands, of users occupying the same virtual environment. Games like Eve and WoW have trouble keeping up with really big groups in the same area and one single planet with zero walls would be a problem for the most powerful computers we’ve ever built. You could be clever and find ways of simplifying the underlying processes (giving up a lot of the complex rendering to the client devices, for instance) but almost all ways of simplifying it eat into the richness of interactions available. So, no sword fights in real time. Not in the middle of the street, anyway.
JanusVR and Lucidscape are just two of the many companies trying to get a working Metaverse off the ground. Lucidscape has scalability at the forefront of their mind, but have gone quiet lately. Hopefully because they’re busy. Janus is probably the closest thing yet to a realization of Neal’s concept. That said, Second Life is very, very close as soon as you factor in VR support. The scaling is going to hold that back, though.
Displays have only just reached sufficient DPI
Until 1080p phones, the DPI of most screens was around the 150 mark. This looks like a game of twister when you hold it against your face. Now we have 1440p phones and that’s in with a real shout of giving you moments of ‘presence’, where you feel as though you are ‘really there’. Palmer Luckey said he thinks we need 8K screens per eye to eliminate the awareness of the screen itself, but we’re nowhere near that and I don’t think that’s preventing us from having very immersive experiences anyway.
Fooling the eyes is the single most difficult aspect to get right and requires the development of sensors and processors that can accurately read your head movements in real-time while projecting an image that matches those movements. It’s unbelievably hard and if it weren’t for the enthusiastic last decade of mobile development, we wouldn’t have a resurgence of VR today.
The input devices have not been created
It’s not a virtual ‘reality’ unless we can treat it like a reality. Simply watching things happen is half the battle. Even now, we don’t have a way of collecting input information in the same way as our interaction with the real world. The real world reacts like we’re real, the virtual world reacts as if we didn’t exist. We have to explain every single detail of our interaction and what that should mean to the virtual environment.
A lot of this problem has already been solved by the incredible physics engines in modern games, but the relationship between our position, body weight, balance, muscle strength, reactions/reflexes and the virtual world has to be so accurate and fast that it’s horribly obvious when it’s done badly. It causes confusion, imbalance and a whole host of other things, including headaches and motion sickness. That last one is a really big problem, trust me, I’ve had to rip the HMD off and sit in front of a fan for half an hour more than once!
There are some ingenious solutions to these problems and we’ve yet to see which are most effective, but with so many options, it’s just a matter of time.
So, in short, the answer is ‘nothing’. We’re on it. Any day now.
See you in the Metaverse