Bitcoin mining isn't melting the planet. It's not even using more energy than Christmas lights or wash dryers. The comparisons made with [insert random country here] is dumb, because any worldwide phenomenon has a large chance to use more energy than [insert same random country here].

Even if the base layer of Bitcoin is using energy, one can argue that the functionality it offers (trustless, borderless, permissionless, programmable, digitally-native money) is worth the energy cost.

> It's not even using more energy than Christmas lights or wash dryers

I can't come up with any reasonable measure by which Christmas lights use more energy than Bitcoin mining. The reports that suggest this seem to extrapolate US figures across the world, which looks unrealistic. And they're old enough that they don't account for the switch to LEDs.

Even if a billion households had 50 strings of LED lights each and kept them lit 24/7 for all of December, they'd still use less energy than the annual usage of Bitcoin.

The 2008 paper released by the DOE pegged lights at 6.63TWH per year across all segments in the US. That's 0.02% of electricity usage in the US at the time, which extrapolated is still only 44 TWH in 2008. With 100% LED lights, the usage would be 0.663TWH according to the 2008 DOE paper. I'd assume we aren't quite there, but even so, much more than 2-3 TWH on holiday lights nowadays would be shocking.

> Even if a billion households had 50 strings of LED lights each and kept them lit 24/7 for all of December, they'd still use less energy than the annual usage of Bitcoin.

On the final math, at 5 watts per strand, you'd wind up with 182.5 TWH. Bitcoin at 150TWH-250TWH puts it into the ballpark. Nice.

But we’ve already established that we can do all that without the energy costs if we end up needing a trusted authority anyway. There’s zero need for proof of work if you’re going to trust a central authority to validate the transaction history. What’s the point of such an extremely costly base layer?

I've yet to see a convincing argument why we end up with a trusted authority. Can you point me to which incentive would cause this?

The subject of this thread is reversibility, and how implementing that would involve trusting a central authority (goverment, etc) to force reversals of transactions on the chain.

Per an ancestor comment you replied to:

> Reversibility is one example of a design requirement that undoes the claimed advantages of permissionless blockchains that are meant to justify their inherent downsides.

If we need reversibility on a permissionless blockchain, why have a blockchain? A postgres database would do fine. You could even broadcast the write-ahead log to allow any party to verify that history isn’t silently rewritten.

You even said it yourself:

> Do you not want to transact on the base layer because of lack of reversibility or fees? Move to a higher layer solution that sacrifices some characteristic of Bitcoin to enable something new.

If we’ve established that reversibility is a table stakes requirement for a financial system, the thing you’re “sacrificing” is the very thing that makes PoW essential. Since PoW is planet-melting, can’t we just use a database instead?

> because any worldwide phenomenon has a large chance to use more energy than

It's not a worldwide phenomenon. There are orders of magnitude more Christmas lights and wash dryers than mining servers. And yet...

> Even if the base layer of Bitcoin is using energy, one can argue that the functionality it offers (trustless, borderless, permissionless, programmable, digitally-native money) is worth the energy cost.

The idea that Bitcoin (#23) is worth as much or more as all the economic and general human activity of Poland (25), Thailand (24), or Vietnam (22) is highly suspect:

* https://digiconomist.net/bitcoin-energy-consumption/

> it’s not even using more energy than Christmas lights

This claim appears to be untrue -

https://charlesarthur.medium.com/does-bitcoin-use-less-energ...

tl;dr - the claim appears to be based on a combination of pre-LED figures in the US, and just plain wrong figures from other countries, and is probably out by a couple of orders of magnitude.