It seems like we are having two different arguments. My main point here has been that material scarcity will likely be a non-factor in the future. You seem to be in agreement with that point, but seem dead set against the practicality of space mining.
Throughout my comments I've maintained that molecular assemblers/disassemblers (and possibly VR) should drastically lessen the amount of materials we consume. I'm simply saying that if we do run into scarcity issues, worst case, space mining should be many, many orders of magnitude cheaper within this time span.
Right now the US goes to the expense of floating rare earth elements on freighters from halfway across the globe, simply because we don't want to deal with the pollution, real property rights, safety issues, regulatory issues, and eyesores resulting from US mining operations. With the expected reductions in cost to achieve orbit, in 2112 the asteroid belt could very easily be the new China.
Now you could argue that the same breakthroughs that allow cheaper space mining would allow us to cleanly and efficiently extract resources in greater number from the earth - "molecular mining", if you will. That may be the case, but it's really impossible to predict to that degree of specificity at this point in time. We're probably arguing over what in the future would be the equivalent of going to the Walmart down the block or the Target across town to get a package of batteries. It probably doesn't matter.
Besides, the amount of materials we need really depends on how ambitious humanity wants to get:
http://en.wikipedia.org/wiki/Matrioshka_brain
