There's an article on the BBC about a guy who's 'invented' a camera that, instead of printing a picture, prints a little description of whatever you just took a picture of.
Of course, we don't actually have that technology yet - it's really sending the photo to amazon's Mechanical Turk, where a human is describing the picture and sending that information back. (A spot-on perfect use for mechanical turk really!)
What interested me, though, was how much the sample description sounded like something you would encounter in an old text adventure. The 'inventor' seems to see this as a downside, wanting to coach people to put a little more humanity into their descriptions. I don't - think of the opportunities! You could set up *real world* text adventures, with players in helmet cameras able to see only a text description of what they were 'looking' at, but the freedom to look and move in all directions... well, carefully.