The big lesson from Second Life to me is that it's a novelty for 95% of potential users, and a fixation for a few true believers.
VR and AR are in that era of radio in the 1920s, or personal computers in 1977. They're interesting, people might gawk at one for a little while if given access to it, but right now, the long-term audience is going to be primarily enthusiasts who are passionate about the technology for its own sake.
We're still waiting for a lot of details to snap into place to make it broadly appealing:
- The hardware and setup needs to be turnkey. Newer kit is getting a lot closer, but I think it's going to be hard because you have to factor in things like "setting up a wide enough floor space to avoid injuring yourself when using it" and "we haven't really resolved that this gives a fair number of people violent sickness"
- There need to be killer apps. Some of the VR experiences seem like they'd be fun, but eventually exhausting. It's sort of like the motion control (Wii/Kinect/PSMove) trend-- people enjoyed them, but it seemed like it burnt through quickly, rather than becoming a core part of new gaming experiences going forward.
AR likely has an easier road to "killer app" because it can be applied to a bunch of vertical use cases; I'm picturing a fry-cook with a heads-up display that tracks how long each patty has been on the grille and its internal temperature, for example. Even if mainstream consumers never buy AR gear, there might be a million devices sold to businesses. Makes me think of Windows CE; the consumer launch was muted, but it was on a billion scanner-oriented devices for years.