Low transaction fees are certainly not the main reason for Bitcoin adoption - the main reason is having a highly fungible (= worldwide exchangeable across borders) store of value that is not subject to inflation or other methods of expropriation by a central party. The core paradigm which ensures that is decentralization.
That's the reason I came to Bitcoin. And all Bitcoiners I know came in because of the same motive: To be freed from the bonds of the terminally ill financial system we have today.
I did not come to Bitcoin to be able to do sub-$1-transactions for free. I'm fine with people wanting to do microtransactions, but they certainly do not need to have the same level of security as for reasonable-value transactions. It's outright idiotic to put the core paradigm of Bitcoin at risk (which is the very reason why Bitcoin has value at all), just to satisfy a spam-for-free use case. Solutions for microtransactions are in development - these transactions should take place in abstraction layers.
ya.ya.yo!
That part is fair enough to a point, but even if we take microtransactions out of the equation, there will still come a time in future where our userbase could easily be large enough to fill a 1MB block on a regular basis. Decentralisation is clearly important, but at the same time, you can't expect 10 minutes worth of transactions from all over the globe to fit neatly into something smaller than a floppy disk forever, right?
Of course some people had problems, but that's a non-argument. If blockspace is limited, people will learn and simply pay higher fees. Bitcoin is simply not suited for buying single lollipops and expecting instant confirmation - such transaction dust is better handled by an abstraction layer that reduces the data footprint and improves confirmation times.
So nothing is needed asap. We need a well-thought-out solution that scales well, not some quick-fix that kicks the can further down the road and at the same time exposes Bitcoin to centralized control.
ya.ya.yo!
This is where I start to disagree because it's a slippery slope. Even if it's only a relatively small amount, say a dollar or two, if a fee is included, it belongs on chain. That's why the fee is paid. We shouldn't view problems with delays on these types of transactions as a "non-argument" because I don't find it likely that people, having experienced a delay in their transaction, will think "
that didn't go well, I'll reward this bad service by paying more next time". They're more likely to arrive at the conclusion "
that didn't go well, I'll use something else next time". If a viable solution isn't found before the blocks start to fill on a regular basis, users will migrate to a service that does confirm their desired transactions in good time and that means there will be less fees collected for miners (not more, like people keep suggesting for reasons that make no sense). I don't hear any users of any other coins expressing this view that their coin should only be used for large transactions and those coins are the ones who stand to benefit if we start pissing off a sizeable chunk of our userbase. If we force users out by turning this open network into a restricted one, there will be consequences to doing that. Once it's done, those users will be difficult to win back.
If Bitcoin ever turns into a niche service that a small number of people use once in a while to move a sizeable amount, that will generate
less fees for miners than millions of people using it several times a day for small and medium sized transactions (and the odd large one). I don't see how this isn't obvious to everyone. There is no gain whatsoever in artificially limiting our target audience. Further, as the block reward diminishes over time, those fees will become more crucial to incentivise miners, so if the network isn't generating large amounts of fees, only the miners with the lowest overheads will be able to afford to mine and that will lead to centralisation. More users equals more fees collected and a larger incentive for more (unique) miners to secure the network. We can't start dismissing all smaller transactions as "dust" and assume they're all worthless. If a fee is paid, enough of them added together will sustain the network and ensure its longevity.