My analysis of Ethereum: It's economically limited. Ignoring the implementation, we could conservatively say that some interesting web applications are convert

My analysis of Ethereum: It's economically limited. Ignoring the implementation, we could conservatively say that some interesting web applications are convertible into the "Dapp" form. Right now, these applications are presented as free to the user(free sign-up, free service) and the cost burden lies almost entirely on the service provider, while on Ethereum the user must pay some of the computation cost of their own transactions. Why is this a problem? Surely everyone can have their personal devices mine and produce ether and then the cost of everyday use is hopefully negligible. But even if that were plausible, there's enough of a power differential between personal devices and large server farms that the latter is going to have a huge cost advantage. From this advantage market forces will seek ways to leverage out a monopoly within the Ethereum platform - to enable spammers, extract tolls, mine for personal information, or any other conceivable mechanisms of earning a profit by crushing competition. It's the position Bitcoin is facing now with the mining pool monopoly. There will be a market and it will have a price, but it won't be a huge, global one, not if it's an essentially centralized system doing centralized things underneath an inefficient blockchain layer. In summary, currency design is hard. Proof-of-work mining basically favors capital pooling by letting the richest buy their work - although they may add some measure of anonymity, the resulting power dynamics are identical to those of gold. Changing the proof metric to another computationally-focused one only changes the cost basis, not the bottom line. As such I'm very sour on proof-of-work today as a decentralization tool. I would consider proof-of-stake or other metrics to be in a fundamentally different category, although they also need answers for the distribution problem.