Um, is it just me or are they really SUPPORTING the network with the fees they are paying on transactions? In 16 - 20 years this is going to be our bread and butter tx fees will likely hit 25+ bitcoins per block within that time frame if my math is even close to correct. (not showing my math here, because it is filled with a lot of conjecture and guesstimation and I really don't feel like having it picked apart.) If we think competition for blocks is harsh now. Just wait until we are looking at 10-100k tx per block and fees that are 5x+ what the generated coins are, also consider that bitcoins will likely be in the $100-$1500 range by then (likely higher) as long as the network stays strong. The adoption is there now. We just need to keep stability in the network. We need to get this shit figured out with larger blocks, because eventually with fees being the major source of income, no one will want to LIMIT block size.
Miners will certainly want to limit free transactions, since they cost money to store. There is also the argument that with unlimited block size we will have a tragedy of the commons situation where everyone uses minimal fees. If there is a limit, transactors will compete for the available slots, driving up fees. It's not as simple as you seem to think.
The fees and block limits discussion is long past due, they are intricately connected. Tighter limits on blocks will lead to more competition for scare resources and thus higher bids (fees) paid to process transactions.
User pays.
Expanding block limits to ridiculous sizes will ensure that blockchain maintenance becomes a rich man's only game .. i.e. only Google, Facebook and Citigroup can afford to maintain full blockchains. Conversely it will keep fees relatively lower for longer .... decisions, decisions. But better we are kept aware of the impending changes before being rushed into them by an employee of Google.
there also exists the "amicable separation" solution and to split into a little persons chain BDB and big guys chain levelDB?