Only just seeing this thread, several pages into its life (and yet to check whether the following is an oft-repeated meme that I oughtn't to bother with), I envisage 3D printing as providing things for which there are no easy 'conventional' manufacturing techniques.
It'll still be easier to create a bolt out of a bar of metal, cut to length and the screw-thread carved out in the current mechanical way, because 3D printing one (even
with suitable 'metallic overlaying' tech, which isn't exactly in the pipeline) makes about as much sense as casting one in a mould through the lost-wax process. You
could do it (with the caveat about the difficulty of making it out of metal), but there's a number of (other) reasons why it wouldn't be as good as a properly machine one.
However, there are doubtless countless uses for shaped items for which casting, extruding, vacuum-forming and other generally-used processes are lacking in easy and cheap (or even
possible) production capability.
You
can make one of those "balls within a ball" things[1] out of plastic by making hemispheres and sealing each around their smaller counterpart, but that way lies imperfections and still some manufacturing difficulties... The ivory ones get carved out of solid material, but that's wasteful of all the material (taken from the holes in each shell, and that was
between each remaining shell). But you could 3D print one
very easily.
Extending that to something
practical, even though it might have ostensibly one-way locking tabs, every container/shell for electrical goods can be prised apart (with care) and put back together. Though it might not please the hardware-hacking (or repair) communities, imagine 3D printing a case around a product, like a much more complex version of injection-moulded plugs, so that there are no seams in the casing.
If we ever
do get to do significant metallic 3D printing, on a suitable scale, think of a safe. The weak-points of the hinges are no longer as weak as in a traditional one if they've been "printed" as part of the whole safe with a perfect same-crystal join between the door and the door-hinge, and the similarly between the body and the body-hinge,
both cast in place so that it cannot even be lifted off (perhaps, although the closed door might be sufficient) and the mechanism within the safe door is not behind a panel (even an internally accessible one) but just
within the mass of the door. Printed in situ, possibly even in a randomly generated orientation (beyond that necessary to align with the keyhole) and surrounded by randomly-generated 'spoiler' bits and/or deadlocking triggers. That could be set up to happen on manufacturer of each item, in a fashion only truly understood by the CAD/CAM-equivalent controlling machine in order to make safes of a complexity to satisfy the most paranoid of users who would rather that even the manufacturer does not hold the full details.
I'm not sure I'd 3D Print a lot of things. Bricks and windows and glass bottles and the aforementioned bolt (plus nuts, screws, etc) all have perfectly good manufacturing processes... The cardboard box of breakfast bars (correction:
previously a container of breakfast bars) that's sitting within all-too-easy-reach is likewise not so easy to see being replaced.
However, the mouse lying (largely unused) by the side of my computer might 'benefit' from being given a shape and construction-method that current processes couldn't as cheaply reproduce. This keyboard could also benefit, and there may be 'better' ways of doing it (hard plastic case and hard plastic keys being seamlessly melded together by a softer plastic skirting
that is just a minor adjustment of the same continuous polymer to mean that there's no gap for crumbs, hairs, sloughed epidermal cells, the crystaline remnants of finger-sweat and dog-knows what else to fall through. And pretty much waterproof, for when you decide to wash the keytops under a tap. Which you
can do with fully "rubber-top" keyboards or "grommited" buttons, but not quite with the same advantages. Whether such a construct would ever break out from (say) space-station or oil-rig locations of use, where a degree of inherent cleanliness (and also capability for in-situ manufacturing!) or resilience against general muck would be considered important, and become a mass-market item... I don't know.
I'm going to re-examine the poll choices before making my final decision (nothing seemed to be quite what I thought, when I first read the list), but doubtless I'll find myself ninjaed, and much more succinctly.
[1]