There's also a lot of people who don't actually care about TVs that much (or maybe i'm the exception? I can't be the only one who just has the TV as background noise/side glances while I use my computer).
My tv is 25 years old, has an RF-aerial input only (have to go through the VCR to play a DVD, and only bought the DVD player 4 years ago).
I still don't even have a digital tuner, and only 2 years before they turn off the analogue broadcasts (which they were meant to turn off end of 2009, but by then not even 25% of aussies could receive digital TV).
But then, my PC monitor is kick-ass (27" 2560x1440 Dell Ultrasharp), and I'd rather spend my money on better lenses...
As for the general population, I'm sure there are a fair few people who would like there to be more 3D TV broadcasts, but it's still a 'niche', and obviously not worth the advertising revenue vs transmission costs. Pay-TV still has some 3D content, but they never even state the fact whenever they advertise on free-to-air. Could be different in the States or Japan, but with the way their economies are going, i'd rather be eating a meal than watching a fancy TV.
Back to photos though, 3D photography has been around for a hundred years or more, I love this guy from 1955-73:
As for digital 3D stills, even the 'canon rep' in some other post said that 3D still were 'inevitable'. But i'm guessing more a 5-10 year timeframe than in 1-3 years.
Just a few musings on hurdles to overcome:
File formats. Are there even any standardised file formats for 3D pictures? (serious question, not rhetorical) Displaying 3D web content like you suggested is certainly plausible, i'd suggest even inevitable (eventually). But think about windows only just having added RAW support to their inbuilt viewers, how long until they add a 3D image-plugin after 3D cameras even materialise? we're just at the point where webpages can play video without flash plugins (newest versions of firefox at least), after 10 years or more of digital video being popular.
Monitors. Early adopters and gamers (and those suckered in by the WK2010 3D broadcasts) might be making inroads to 3d TVs, but that's a very small proportion of people. I know people still using CRT monitors for web-browsing. Tablet PCs and laptops are taking over from the desktop PC, maybe it's easier for them to make 3D screens on the latest gidget, I don't know. But for those who *have* to upgrade their phone every year or more, there has to be new features in the new model (to justify it being a 'new' model), and eventually that new feature will be a 3D screen for which you don't need glasses. And when that happens, the content will come, eventually.
Printing. I know not as many people print photos compared to looking at them on a screen, but i'd guess just as many print now as did back in the film days. Until someone figures out how to print a 3D photo to paper, there's a big roadblock. Of course, there are digital photo-frames (despite the horridly low-resolution of the ones i've seen), they can be made 3D as easily as a TV. But I predict them to die a slow death, with everyone becoming more and more power-conscious and turning TVs off at the plug to save a Watt or two worth of standby power, who'd have a screen projecting an image 24/7?
Cameras. There's your biggest sticking point, imho. Check out the Sputnik above. My main problems with making this digital are:
Two lenses or one?
One lens makes focussing more accurate, but then what? split the image down the middle (if it can even be done, i haven't thought enough about it), and the sensor has to be smaller to cover the image circle, or you need an MF-sized lens. Then the left side of the right image, and the right side of the left image are in the middle of the image circle = good IQ. The other edges, however are (almost by definition) fuzzier on all but the best lenses. Think about field curvature too, no way something like the 24-70 would cut it.
Two lenses, for a start you're limited to primes without some way to ultra-precisely match zoom lengths. Even with prime lenses, think about focussing. with anything faster than f/2.8 and/or longer than 35mm or so, DOF is going to be small. By definition the 2 lenses are looking at different angles, and they might be AFing on slightly different points. Leica may have more success adapting a rangefinder lens for 3D than a canikon AF lens.
One body or two? A two-lens single body will be huge. Huge. You have to make the body wide enough that any lens won't bump into the other. If you take two 400 f/2.8s side-by-side on one body, that body's going to be at least 30-50cm wide, not to mention the weight. Or limit the body to only certain lenses, and you're alienating a lot of people and uses. (I really do like the idea of 3D telezoom photos of wildlife, now that I think about it)
Two bodies? Well, if you can get two well-matched primes and bodies, MF them accurately (and/or stop down) enough, trigger them at precisely the same time, then you can make a nice 3D photo. But then, you can do that now, if you can afford to double your kit. No need to build a new body for that, you just need the file-format and processing software (which probably does exist already).
I'm not saying that a single-body consumer-affordable solution is not going to happen, i'm certain it will happen one day, but i think i'm more than safe saying i'll see a 100mp 1Ds mk7 first.
Or widen the search, with a radically new body, with a new lens-mount, with a dedicated non-changeable lens, with lower quality pics, it could happen as soon as a year or two, at least as a prototype/gimmick...