Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

SatoshiGuesser – Roll for Bitcoin

github.com

Post reply on HN
Print view View original

51–57 of 57 posts

Re: SatoshiGuesser – Roll for Bitcoin

#51

Post by thaumasiotes »

CobrastanJorji wrote: post #25
ivanjermakov wrote: post #20

Earlier quoted context omitted.

That's gonna be a nice storage bill!

Let's see...2^241 or so possible 256 bit numbers, so that's 256 * 2^241, so that's....10^50 yottabytes. Obviously we're gonna need cloud storage for all this, so let's say that's about 2 cents per gigabyte/month, so that's...2.2614 × 10^63 dollars per month? Actually, why does the site list the odds as ~1 in 5.27 × 10⁷²? That's 2^241, but it's picking random 256 bit numbers. Is it because there are so many valid hits…

> Obviously we're gonna need cloud storage for all this

You can keep a comprehensive list of "all 256-bit numbers tried so far" in 256 bits of storage.

Re: SatoshiGuesser – Roll for Bitcoin

#52

Post by sunrunner »

thaumasiotes wrote: post #51
CobrastanJorji wrote: post #25

Earlier quoted context omitted.

Let's see...2^241 or so possible 256 bit numbers, so that's 256 * 2^241, so that's....10^50 yottabytes. Obviously we're gonna need cloud storage for all this, so let's say that's about 2 cents per gigabyte/month, so that's...2.2614 × 10^63 dollars per month? Actually, why does the site list the odds as ~1 in 5.27 × 10⁷²? That's 2^241, but it's picking random 256 bit numbers. Is it because there are so many valid hits…

> Obviously we're gonna need cloud storage for all this You can keep a comprehensive list of "all 256-bit numbers tried so far" in 256 bits of storage.

If Advent of Code has taught me anything it’s that interval ranges can be really useful for this kind of thing. I mean at least twice in ten years. We just need to figure out how to coordinate individuals attempts to make it storage efficient.

Re: SatoshiGuesser – Roll for Bitcoin

#53

Post by sunrunner »

sanswork wrote: post #40
MattCruikshank wrote: post #35

Quick question - why hasn't someone 51 percent attacked Satoshi's wallets? Estimated cost of a 51% attack on Bitcoin, if no one is cooperating, is $6 billion to $10 billion. Surely the cost goes down if they get some big players to cooperate. And the reward is... $83 billion. Basically 10x your money. I mean, this is the kind of thing that we could sell bonds for, to raise the $6 to $10 billion needed. Other than the…

51% attack on bitcoin doesn't let you remove coins from someones wallet it allows you to change which transaction history is considered the real one. So you could send someone bitcoin then do a 51% attack to make the chain without that transaction longest so you get to keep your bitcoin but you can't use it to just take money out of someone elses wallet.

> doesn't let you remove coins from someones wallet it allows you to change which transaction history is considered the real one

Potato potahto.

“You’ve got it all wrong officer, I wasn’t talking money from his wallet, I was changing history such that the money was transferred to my wallet instead.”

Re: SatoshiGuesser – Roll for Bitcoin

#54

Post by meowface »

futune wrote: post #42
meowface wrote: post #33

I'm not opposed to LLM-generated code at all, but the such obviously LLM-written README is annoying. The style is so easy to spot. At least try to figure out how to prompt it to not write so obviously like an LLM. (And no, I'm not even referring to the em dashes.)

Why would you want LLM content to masquarade as non-LLM content? I think I'd rather have it be obvious if people are going to be using it anyway.

A synthesis here is to include a "Written by Claude/GPT/whatever" at the bottom but still make it read non-slopfully.

Re: SatoshiGuesser – Roll for Bitcoin

#55

Post by futune »

meowface wrote: post #54
futune wrote: post #42

Earlier quoted context omitted.

Why would you want LLM content to masquarade as non-LLM content? I think I'd rather have it be obvious if people are going to be using it anyway.

A synthesis here is to include a "Written by Claude/GPT/whatever" at the bottom but still make it read non-slopfully.

That does seem like a prosocial approach.

Re: SatoshiGuesser – Roll for Bitcoin

#56

Post by sanswork »

sunrunner wrote: post #53
sanswork wrote: post #40

Earlier quoted context omitted.

51% attack on bitcoin doesn't let you remove coins from someones wallet it allows you to change which transaction history is considered the real one. So you could send someone bitcoin then do a 51% attack to make the chain without that transaction longest so you get to keep your bitcoin but you can't use it to just take money out of someone elses wallet.

> doesn't let you remove coins from someones wallet it allows you to change which transaction history is considered the real one Potato potahto. “You’ve got it all wrong officer, I wasn’t talking money from his wallet, I was changing history such that the money was transferred to my wallet instead.”

It's not about if one is wrong or not it's about what you can do with it. With a 51% attack you need to have a recent transaction to the wallet you want to attack and have received economic benefit from the transaction already.

It allows you do to a refund it doesn't allow you to take random bitcoins from a wallet.

Post reply on HN