This is precisely the problem that Bitcoin was created to solve. You can argue that Bitcoin specifically has failed to fill that role; but, I still earnestly believe that a cryptocurrency system will be a key aspect of the "better web", that the crypto infrastructure is in its infancy now and will bloom as development continues.
2 certainly requires state involvement. it is impossible to use technical means to solve a social problem, since bad actors will just find the loopholes which can only be closed after the fact. In a criminal court, there aren't so many loopholes and they can be challenged in real time.
I think a platform like Flattr makes sense: you pay a flat $X/month to Flattr, and they handle the logging of what you watched/read, and splitting up your $X among those creators.
The solutions you mention all work just fine--lots of artists are making money on Bandcamp and Patreon. The only reason you think they don't work fine is that they haven't scaled well for gigantic centralized entities. But I have to ask, why, as a user, would we want gigantic centralized entities? That's the absolute antithesis of what I want.
What is valuable on the web?
Thinking more about this, one of the only tangible currencies on the real web is hosting/processing capabilities: you can measure it from the web itself.
I think that's what Bitcoin tried to achieve, although it does so without providing real value to anyone. Instead, imagine the following: you provide a slice of your computer time to Google, some local storage to blackblaze, some bandwidth to Netflix.
All of these are commodities you pay for in the real world. They are valuable on the Internet. Lend some storage and bandwidth to youtube, they could credit your account, and you could use that virtual currency to spend on youtube creators. Youtube would directly pay them with $$$ instead of paying you.
However, my answer only mentioned huge-scale companies, I am not sure it would make a lot of sense at a smaller scale, unfortunately. However, that basic idea of service exchange is something I've been thinking for a while as a path forward to a new kind of currency, which I should put into a prototype as soon as I have the time to do so :)
This one is a bit easier. Prepaid credit cards are frequently accepted online and can be purchased with cash. The fees are a bit high as is but there's no reason merchants can't create more of these and likely even lower the prices.
Alternatively your CC provider could potentially act as a privacy guard by providing randomized rotating onetime transaction numbers as well, I know some pay services like Samsung pay do this to some degree already by creating virtual cards that require a third party to authenticate without giving your actual account, though I believe those virtual cards remain fairly static. With more potential entropy (large GUID), it seems reasonable you could rotate those on the fly.
This is largely down to "Know your Customer" laws and the concern they will be used for Money Laundering
but stuff on the web is not at all like stuff in the real world!
this is a hugely understanted problem. digital artifacts have zero reproduction cost (beyond a few volts and a couple of amperes, which are provided by whomever copies)
therefore, IMO, the market mechanism we have in place, which evolved to optimize material goods, is woefully inadequate to deal with digital "goods"
finally, this is not a technological problem, but a sociopolitical problem.
I am with you on this. With the cash equivalent for the web I meant more the following: frictionless, commission, private transactions that are native to the web in the way cash is native to the real world. The economics of it dictates a cash-replica won't do.
1. How to pay for stuff in the web in the same way you can pay with cash in the real world.
2. How to regulate the new-age, digital-good, information-aggregation monopolies. I suspect this will have to be done by either a state-forced, or a highly-useful interoperability protocol for building new tech.