But what is the REAL PROBLEM ?
The real problem is the remuneration of the miners when the block reward runs out, and the game-theoretical aspects that follow from that. Without a block size limit that creates artificial transaction scarcity, miners cannot force users to compete for a scarce resource, and will not get paid enough to secure the block chain with the silly PoW security. Moreover, without fees, the block chain room becomes a resource that will be wasted by spam or many parasitic applications.
The problem that Satoshi didn't know how to solve and which has no good game-theoretical solution in the frame of the (silly) economic rules that Satoshi gave to bitcoin, is: how to secure the chain when the block reward runs out ? THIS is the real problem. Nobody knows how to solve it in a satisfactory way with PoW and sound money theory.
This is what will kill bitcoin in the end. Not the mumbo jumbo about resources or decentralisation. These are straw man arguments to keep people away from the true problem.
Sorry, I believe this problem has been thought about and discussed quite a bit, though it is far enough away that not much serious action has been taken.
Though I'm not ready to run it, I'm actually intrigued by Bitcoin Unlimited's approach of letting the miners vote on the size of the blocks. Since the miners benefit from transaction fees, doesn't this effectively give them (collective) control over the supply of block space? They will increase supply when the fee market is too small, to increase their customer base. They will decrease supply when the fees are too low, to increase scarcity and drive the price up. At some point in the middle there will be an equilibrium where the fees are high enough to pay the miners' costs (and very little if any profit, otherwise more miners would join the market), but low enough that users will want to use the network.
Is there a hole in my thinking here?