It is most certainly a reasonable consumption for a payment system. Every society with inflationary money eventually grinds to a halt and crumbles. The world must absolutely have the best implementation of money that it can get its hands on.
Energy use is not its defense against attacks. Distributed hashing is. The hashrate can be much lower and Bitcoin would still work. The difficulty would adjust downward.
Obviously, the hashrate will top out at some point in the future. From that point on, it can dwindle and that won't be a problem. Just so long as a certain minimum amount stays around to keep the network running. That won't be a problem in a future where Bitcoin has become the new global currency.
Re: "So perhaps people should quit criticizing bitcoin for its energy intensity and start criticizing states and nations for still providing new industries with dirty power supplies instead."
Three responses:
(1) The author has introduced another unrelated claim, presumably to distract from the main point. See the informal usage of "non-sequitur" at https://en.wikipedia.org/wiki/Formal_fallacy
(2) This is a false choice. The two arguments are separate.
(3) There are better alternatives when it comes to energy efficiency: (A) in the current cryptocurrency space; (B) in the financial transaction space; (C) in the (open-ended, what might the future hold) software design space.
...
The article misses the point. If we're talking energy, the main criticism of Bitcoin is its efficiency. I mean "efficiency" as defined in the engineering sense: useful output / input. [1]
If we define useful output as one Bitcoin transaction, how much input energy does that require?
According to [2], "Because of all that calculation, the energy cost of Bitcoin is high in comparison with that of conventional financial transactions. For example, according to one estimate, processing a bitcoin transaction consumes more than 5,000 times as much energy as using a Visa credit card."
According to [3], "Bitcoin's power consumption is extremely high compared to conventional digital payment, and one transaction now uses as much energy as your house in a week"
Finally, here's my personal perspective. I have relatively little confidence in a simple, generalized idea of cryptocurrencies. As far as the basic computer science, I like blockchain technologies roughly as much as I like Merkle trees. I'm also open to see where alternative forms of transactions can go.
That said, these forms have design constraints that the technologies must address. I would like to see particular innovation in the design of particular cryptocurrencies that make them suitable for particular uses. For a system that hopes to scale up transactions for a broad user base, I think per-transaction energy efficiency is an important metric.
As if using 2% of all energy consumption is reasonable for a payment system.
And at 100x current market size, it'd still be hardly used.
And it can't really start using less energy as it matures, because its insane energy use is its defence against hostile network takeover.