While the block reward is still in place there are many reasonable solutions that are "good enough". For example the quadratic penalty function in Cryptonote (Monero) or even relying on the probalility of orphan blocks to create a fee market that deters spam. The difficulty is that this blocksize debate has raised an even more fundamental question: How does one secure the network after the emission runs out? Monero has solved this with a fixed tail emission, Dogecoin also has a tail emission and so do many POS coins (Peercoin for example); however this is a non starter for Bitcoin because of the social covenant. Furthermore a fixed blocksize may even not provide for a fee market to develop, in the absence of a block reward, as the current "stress tests" are demonstrating. The small block crowd are by the way also making some very valid points.

The best I have seen to date is BIP 100 with a 50% threshold, and much higher starting limits; however this is dependent on external market factors (other crypto currencies, fiat payment methods etc.) to keep the "mining cartel" from raising the fees with no limit. I have been researching this issue for over three years and unfortunatly the more I look into this, the more discouraged about the future of Bitcoin I become.

What is needed here is a solution that allows for a dynamic blocksize limit that will respond to market forces allowing a fee market to develop, in the absence of block rewards. I am not saying that such a solution is not possible, I have simply not found it. Pending a solution, a bear I remain.