Earlier quoted context omitted.
Seems wasteful & fruitless. Better to evolve a micropayments model that works (because at least i can quantify what I pay)
In principle, yes, PLEASE!!! In practice, people have been wishing for a well-functioning micropayment system for well over twenty years and the best we have to show for it is PayPal. Flattr seemed like a good idea, mostly, but for some reason it tanked, too (AFAIK). Either people are not interested in creating one that works, or the problem is really, really hard.
I think the problem is simply hard. Micro-payments imply more infrastructure costs (more transactions) but lower revenue (because the payment itself is tiny, so the fee has to be tiny as well).
But look at Bitcoin's Lightning Network, it may very well become a valid micro-payment system. As far as I know it's biggest scaling roadblock is getting efficient routing.