Let's remember:
Not much amateur (or even enthusiast) is ever viewed on a screen larger than 50', and a lot gets viewed on YT at the default 480p resolution at around 3mpbs, using only a small box on the screen, since higher rates cause many people's connections to gag and require buffering, and the content probably doesn't warrant the added wait or trouble.
Theater audiences generally flock to action, kiddie animations, or romantic spectacles that don't benefit a lot from higher resolution, although I'd bet that portions of the special effects sequences of "Hugo" and similar works were done in labs at 8k, then downsampled to 4k.
The advantages of downsampling are purely notional. The HF G10 is probably the best prosumer videocam, yet its sensor is only 2k. Large pixels and less downsizing give it better low light performance.
No one will want to buy a 4k screen if it costs 4X a 1920x1080 screen and there is no 4k content besides upscaled 2k.
I'd be tickled to see a side-by-side comparison of FF DSLR video against that shot with an HF 610, a m4/3 camera, and the $5k JVC 4k videocam. Zacuto could probably do this, but won't bother, since its target clients have money to spare for fancy stuff, and there's no gain in showing how you can attain 90% of the IQ for 10% of the cost of a RED.
More theaters or public conference and exhibitions may invest in 4k equipment, but not have much 4k content to display, and audiences may be more interested in the quality of the beer and appetizers served, as they browse there iPhones, during the show.
The idea of shooting in 4k so that one can crop when editing, as can be done all the time with still photos, is appealing, but takes a strong PC, lots of time, and mastery of the pro-grade editing software products. Meanwhile, most people never edit video at all, and lots of stuff posted at Vimeo shows only rudimentary editing.
