And I would claim item number 2 doesn't happen often enough to justify the costs of 3D printing for most people. A new handle is less than $2 at Home Depot and takes 20 minutes to go buy. A 3D printer is a several hundred dollar investment, the printing material is another big cost, and the printing time itself is longer than it'd take to drive to the store.
Nobody can deny 3D printing will be great for prototyping and some kinds of custom manufacturing, but I've yet to see a convincing argument for consumer 3D printing.
(3) Also doesn't presume anything. Yes, most people do not have CNC machines (although I know several who do), but you might surprised how many people have wood/metal shops in their garages or basements. One of the other comments on this very story has someone talking about how having a buddy with a metal shop and a dump is just as good a FDM machine.
Again, (4) has absolutely nothing to do with setting. How does being able to cheaply whip off multiple variations of a model having anything to do with industry? Aside from the fact that such methods are often used in industry, because it's a smarter way to do things. Most home-hacking happens one part at a time for exactly the reason I mention.
But I think that's a pretty small group, and I don't it's big enough to justify all the overblown talk of 3D printing revolutionizing the world in general.
Current 3D printers are not revolutionizing the world yet, but the next desktop factories will.
I've used several different models of 3D printers, from the cheapest, to home brewed, to industrial grade, and even the fancy UV cured resin ones. Honestly they really are great for a number of purposes:
1. Quickly getting a near-full fidelity part to hold in your hand. Whether you're engineering something or doing something around the house, there is no substitute for having the thing in your hand.
2. Fast "create and forget" replacements. Handle on your workbench broken/missing/shitty? Print one that fits, screw it on and forget about it. No driving to the hardware store, browsing around, paying, "oops the screw holes are too far apart" etc. How many times have you been making something out of wood at home and only gone to the hardware store to get a handle/hinge/pin?
3. Making 3D parts that would be a huge pain, or impossible, to machine by hand or via cnc. Really, anybody who has spent some time machining parts will tell you that there is more to making a decent part than just designing it and hopping on the mill. Do you have the right tooling? How are you going to hold it? Do I have enough material to hold? What speeds and feeds do I need to use? CNC machining adds an entire extra layer of CAD, and then you still have to worry about most of that stuff.
4. Making several different versions of something, because you aren't sure which one will fit best, look nicest or suck the least. It's pretty rare to see people machine (or even wood form) 5 parts that are identical except for the position of some groove. It just takes too much time, and it's boring.
Don't get me wrong, there's still tons of room for improvement for additive processes, and sometimes a mill or a table saw is a great tool for the job, but enough with this "overbuzz" business. No, you will not be printing out CPUs any time soon, but that doesn't make 3D printers all "buzz".
EDIT: Formatting.