The difference between your hack, the many existing head-mounted displays and the Rift is how serious they're taking a) the FOV and b) latency. These two factors are critical to making the VR feel like something more than a display strapped to your head. This is also why I don't see this working with the PS4 very well either, unfortunately. For the Rift and, really, any VR, to work well you need high resolution, high framerate, and low latency which is not something the consoles are very good at.
As for content, I think this type of VR proposition is way better off than 3D TV/video stuff, as most 3D games can easily be adjusted to render the scene twice with a slightly different camera position. You don't really even need the official dev support for this, as this is something that can be done on the driver level for any DX/OpenGL game.
I'm not really sure about the Oculus API policy but Valve hasn't been particularly "open" about anything either so far. In any case, although the Rift might start off with a single model, that doesn't have to be always the case, it just makes sense to start with one unified offering rather than immediately confusing the market with a bunch of versions.
Anyway, what makes it difficult for me to predict how this goes is a fundamental part of the experience - you have to strap something to your head and isolate yourself completely from the real environment, which is a rather dorky proposition that not everyone would accept. Anyway, even if none of the mainstream games support the Rift/Valve VR, a high resolution and low latency HMD is something I'd definitely want anyway.