A year ago I suggested that if increased capacity were urgently needed we could just do a 2MB hardfork-- which, while causing harm, wouldn't likely cause irrecoverable harm. It was aggressively opposed by the persons demanding effectively unlimited blocksizes.
https://np.reddit.com/r/bitcoinxt/comments/42ftps/list_of_reputations_that_core_supporters_have/czcu3sw
So #rekt.
Yawn. Wake me up if there is an "urgently needed" capacity increase. I'm not expecting one for at least a year or two...
So long as competitive fees effectively pay for tx priority within Bitcoin's fee-for-service markets (plural because in- and out-of-band markets exist), Bitcoin is just starting to work the way it's supposed to.
The idea of FREE TX 4EVA is bunk; block subsidies won't last forever and we must have a better, loop-of-self-sufficiency-closing way to pay miners for our collective security.
If fee markets fail to develop as block subsidies decline, there will be less incentive for miners to secure the One True Blockchain. And that would lead to talk of a tail emission, which might entail raising the 21e6 coin limit.
Do you really wan to go there?
If you want free tx, use a less popular, alternative chain with (effectively) no minimum fee.