For what it's worth, VR has been trying to become a thing since the mid-80s. Much like 3D, it bubbles to the surface for a few years and then nobody cares again for the rest of the decade.
Carmack is clearly an amazing programmer, but I'd like to know what his prognosticatory credentials are. He's also predicting Haskell will be a big deal in gaming. You can be amazing and still be a crackpot. Look at Linus Pauling for instance.
I could have the equivalent of six 30'' screens in front of me; I could switch between them just by turning to look at them - all whilst actually only using a laptop on the sofa (or on the train, but I might look a little odd doing it).
To be good this would require the Rift to have reached a resolution where I can't see the individual pixels - the '4k resolutions' mentioned in the article perhaps. Even if such a device cost $1000 it would still be cheaper than all of those screens.
I can honestly see such a thing being tremendously useful for people who do CAD, deal with big architecture diagrams etc... People who generally like to use a lot of monitors for day to day work.
One application is therapy, for example to combat phobias. So you show someone with a fear of public speaking (to the extent where they have panic attacks) a room full of people and let them speak. This is already in use, but mostly done in front of monitors.
No, not world changing but still very cool.
Of course, in time something like holograms will be more relevant, but in the mean time I think things like Oculus Rift in combination with other technologies can be a nice stop-gap.
For all the familiarity people have with the abstractions of interfaces on 2D screens, they're still just a flat limited surface within the space of our visual perception, and required skeuomorphism to ease people into familiarity therewith. VR is a massive change, which has to start by satisfying familiarity via sufficient imitation of the real world; once someone gets comfortable using it to the point of foregoing real scenarios in favor of virtual ones, then we can start tweaking the virtual to do things previously impossible and yet are natural extensions of their new understanding of the VR interface.
One might further theorize that the design profession itself might be fundamentally changed by it. When you think about it, the mouse and keyboard isn't an ideal medium for this type of design; it's a crude interaction. Why look at a scaled wireframe of a structure when you might be able to virtually build a structure while being inside it - using your hands and eyes, like clay. You could use your arms to gesture, saying "I want this wall here, and to be this high"; "I want a 3'x5' window on this wall, here", etc. You could mold structures with your hands with this type of virtual reality and some type of motion tracking. It would be ground-breaking.
It might change our actions as people and as societies. Other mediums like film, television, and the Internet certainly have.
I don't think immersiveness is necessarily a desirable property in entertainment. Movies, for instance, seem to be less immersive when they're more realistic (the Hobbit movie, for instance). There is something eerie about daytime soaps that arises partly out of their higher framerate. McLuhan did say something to the effect of movies being a "cool" media, meaning you could be absorbed into them. This characteristic seems to be inverse to higher detail. If I were to make a bald-faced guess, I would mumble something about higher sensory refinement taking processing away from appreciating the other aspects of the art. I don't think it's impossible that the fidelity we have today for film is essentially perfect as far as our perception and enjoyment goes--that is, after all, the case for music. This might sound like a controversial opinion here, amongst people who have inherent appreciation for the technical, but amongst our families I bet it will sound more plausible.
That said, I don't know what percentage of people will want to become fully immersed in a computer world. It can be disorientating when you put on the headset and the real world gets replaced by a computer generated one. I think the majority of the world probably won't jump on board until convincing augmented reality becomes possible. I kind of see VR as a building block for AR since they share a lot of the same problems.
There is a pretty big segment of industry out there for which more realistic simulation of the environment without the need for specialised environments (the aforementioned mining equipment simulators were literally small rooms) would be a huge bonus.
As an educator sure, more immersive stuff for things like geography, history etc would be nice, but I think the training (vs teaching) aspect is more interesting.
It will have a disruptive change to more than gaming. Here's why I believe so:
If you look at human history, we started with writing, then newspapers, then radio, followed by television. The way in which we consume media hasn't changed from television/displays. Virtual reality is the next evolutionary step in media consumption. The final step being human-computer interfaces.
I think because it enters us into a new realm it will be disruptive.
All of the examples you mentioned had applications besides entertainment, and in most cases those applications came first. As far as I can see, VR does not have any application outside entertainment. Maybe we're at a point where an innovation in entertainment is "sufficient" to disrupt things, but it's not safe to assume that making something more realistic is going to make it better. The 3D movie/TV craze is greatly diminished today compared to a couple years ago. There shouldn't be anything surprising about that--Hollywood has been trying to get people excited about 3D movies since the 50s, but the simple reality is that we are not insufficiently entertained without it.
I can't find the exact post now, but this link shows the same technology.
Especially if they get into the peripheral market...