If the userbase increases and blocks are full, the choices are as follows:
- Pay an average fee and still wait for the next available block that isn't full (which will take longer and longer as more users join, even if you pay a fee)
- Pay an above-average fee and hope that it's more than everyone else is paying (and if you don't guess correctly, wait for the next block)
- Go off chain and hope that whoever you trusted doesn't run off with the coins
- Use another coin
Obviously the whales won't mind paying the above-average fee, but everyone else will be forced to introduce risk to their transaction. It will either be delayed, or if you go off chain, you have to place trust in a third party. Guaranteed security for them, risk for everyone else.
The fees are pretty tiny, they aren't what supports mining, the inflation supports the mining. The 20 year proposal bridges the time where this may be projected to change from inflation to fee supported mining. This may make it particularly dangerous in its assumptions. We won't really know until we get there.
Or, we could remove the pointless, artificial 1MB bottleneck and everything carries on exactly like it does now. The blockchain doesn't instantly double or treble in size. It doesn't lead to mass-centralisation (or if it does, it will happen much more slowly than it would if we start forcing users off-chain). None of the other idiotic horror stories your side of the debate has raised will come true.
The limit isn't pointless. Please understand the reasons satoshi added it before progressing further.
When the fork goes ahead, you're going to tag along and play nice because you don't have any other choice.
This sentence does even more damage to your position than anything else you have written here.