Hacker News Reader

Comment by wyldfire | Hacker News Reader

I will posit that the proof-of-work is the only way to have: (1) a trustless decentralized network with (2) equitable distribution/minting of the coin.

I will concede that other coins exist that don't use a proof-of-work (instead proof-of-stake or other). Those forfeit trustlessness and/or equitable distribution. I will also concede that it may be possible for those coins to fulfill a need and be useful without all of those things. Though I doubt they ever could have been created if not for bitcoin having been created first.

View on HN

Explaining your reason for trusting it doesn't detract from the fact that you are just trading entropy around. "Proof-of-work?" What work? I spent a undefined amount of power from the grid to render a string of digits in a specific pattern? I can flail against my keyboard or pipe /dev/urandom to a file is that worth $16k? Can't you see that you are engaging in madness? Just because you have defined the method of your madness doesn't make it any less insane.

It's not madness. I think you're dismissing it as "insane" without understanding some critical issues. It's [proof-of-work] the only way to do that [what bitcoin does], or rather no one has yet suggested an alternative that preserves those qualities.

The work done for the bitcoin network has utility, that much we can agree on? And you think perhaps that the utility provided by the network is not a net benefit considering the cost of the PoW? Well, unfortunately that much is subjective and there's enough folks out there willing to spend resources on it.

However it's not madness. Proof-of-work is required in order to have a mechanism that has bitcoin's qualities.

If you want to fight against what you might perceive as waste or madness, it seems like tilting at windmills IMO. But the best thing you could do is build a better mousetrap -- or invest/promote one of the non-PoW coins out there.

EDIT: clarified some uses of "it" in the first paragraph.

> The work done for the bitcoin network has utility, that much we can agree on?

No, I do not agree, there is no intrinsic value in wasted compute. The only value that wasted compute can have would be a measure of the amount of entropy spent generating that "value." Entropy has no value. It's just energy that cannot be spent again.

Ah, ok, well let me lay out my argument for why it has utility.

Your bank and other organizations (Western Union, etc) spend a great deal of their time and money to create and maintain networks that permit them to move money from one place to another on Earth. Your bank probably pays to some network in order to be able to send and receive "wires" of money to and fro. Migrant workers use Western Union and the like to send money from the country they work to their families back home. For this, Western Union and the global banking networks charge fees. The fees charged by bitcoin to relay funds can be significantly lower. Probably not universally so, but I'd say all transactions above a certain dollar amount would be cheaper. They'll probably be as fast or faster than traditional remittances. Bitcoin does this as good or better. Western Union and friends could decide that leveraging the bitcoin network might actually be cheaper than their current network, or they might be forced to realize this by competitive pressure from other bitcoin-based remittance companies.

Ok, so that is one use that is IMO unassailable. The people on this planet is putting money into building retail locations for Western Union, they're paying for the power to keep the lights on at Western Union, and they're paying the salaries of the employees at those retail locations, and the folks who staff the armored cars moving the actual cash from place to place. That's not a waste, right? It's spent by us collectively in order to get that value. And maybe bitcoin's difficulty-scaled PoW is not spending electricity well. But it's useful and that's why people do it.

Now your claim is regarding the value of entropy and I will claim that you cannot get a mechanism with bitcoin's qualities (trustless + equitable distribution) unless you have a PoW. But I will concede that you can have a delegated-trust proof-of-stake coin that has some of bitcoin's qualities without a proof-of-work.

I see no current evidence to support "The fees charged by bitcoin to relay funds can be significantly lower."

In the world of finance, the effective cost of technical transactions is tiny (<0.02USD for mass straight-through processing transactions that can handle batching and delay, <1USD for real time settlement of large amounts). Western Union and friends won't realize that leveraging the bitcoin network might actually be cheaper than their current network, because it's not cheaper, not even close to it.

The fee for payments tends to be much larger because of everything else bundled with the payment, but an equivalent service that provides the same things on top of bitcoin would also have to charge similar fees. For example, a large part of credit card fees pays for chargeback and fraud insurance, and all kinds of bonuses/miles/etc which customers like, so if a BTC service wants to take over that business, it'd have to compete with that. Bitcoin transfers are irrevocable, but most consumers want most of their payments to be revocable in certain conditions, so a future "buying stuff with BTC" payment will have to include an escrow service for an additional fee. The branches that allow you to hand over your local currency cash for a remittance, those are expensive, so a remittance service that uses BTC will cost the same as a remittance service that does not.

Well let me know when you can created a distributed network where more than 30% of nodes can fail without a PoW algo. Some are in development of course but until Satoshi did it, 30% was the limit, now we're at 51%

Two ways to deal with PoS distribution: either start out with PoW, or do a well-publicized open sale of the initial token supply. From the coin purchaser's perspective, contributing funds to a sale is economically equivalent to paying the electric company.

I certainly agree that Bitcoin was a breakthrough that paved the way.

https://medium.com/@mkogan4/proof-of-stake-vs-proof-of-work-...

There are non-probabilistic solutions to the byzantine general problem. Those could work as a replacement of proof-of-work.

Reply on news.ycombinator.com