Yeah, people have been coming up with these ideas for at least a decade, but no actually effective CAPTCHA has ever been developed. Even if we did, rich people would just have buildings full of slaves doing CAPTCHAs =\
The whole premise of PoW is that more money investment == more mining power, this is just as true for “memory constrained” algos as it is for the current hashing one. So the only thing changing the PoW would do is piss off investors by causing them to reinvest.
I’m not saying It’s impossible for us to come up with a better system, its just that I haven’t seen any compelling replacements.
yes. so instead of competing for electricity, they would start competing for memory cards. so in the end they would use less electricity.
the second benefit is to change the puzzle to take away the pros’ advantage. make the requirements to play so that it’s easy for home users but not much easier for pros.
they would start competing for memory cards. so in the end they would use less electricity.
In general morememory cards == more power, but even if this wasn’t the case, then the price of memory chips/ silicon would go up and it would be even harder to buy a video card =/ The people/ places that had better access to chips would wield power over the network. The nice thing about electricity is there are a bunch of ways to make it, so the playing field is pretty level.
make the requirements to play so that it’s easy for home users but not much easier for pros.
This would be nice, but none of the conceived ideas have been made to be robust. Sybil attacks are seemingly always possible in a decentralized network. There are some mitigation strategies but nothing bullet proof I think so far.
Bitcoins strategy was to rely on greedy investors to secure the network. I think of bitcoin more of as an experiment than anything, so changing it would ruin the experiment. A more socially-oriented network would have to supersede it from the outside for anything to change bitcoin realistically.
That’s interesting about Sybil attacks. I’ve thought about that - my idea was to have a global online convention, once per year, that lasts nine minutes. During that time you must pass a turing test from an existing user, and then give a turing test to two other (new or existing) users. So it would be very hard to obtain multiple identities, without being a real user.
The turing tests can be video calls where you ask each other to draw pictures, on video, of things that have never existed.
Actually that’s not mentioned in the wiki article. Maybe I’m the first person to think of it.
I’ve been thinking a bit about what you said. If the puzzle were changed, so that instead of just competing for electricity and GPUs, they are competing for RAM cards too … how would that affect the difficulty and the strain on global resources (not just talking about electricity now).
You change the mining function somehow, that makes it much much more expensive for pros, but are free for normal people mining on laptops. Some kind of function that requires:
allocating 1GB RAM and 10GB disk space
human input
once a block is made on this computer, it can’t start work on another block for a while
can only mine for 8 hours a day
That sort of thing. The result, I think, would be this:
More people will mine. The difficulty will go up. The share of blocks going to pros will go down. The profitability for pros will go down in proportion with the share of blocks they are getting. Many pros will stop mining. The global resources used for mining will go down.
EDIT:
After not much research, somebody has thought of this before.
The example of a mining function like this (i had to look it up again) is Monero and randomX. Proof of allocating disk space also already exists. And there is also a currency which changes mining algorithm every block, but I forget which one.
It’s an idea a lot of people have been working on. If my logic above is sound, this approach not only solves the electricity wastage problem, but also the hardware wastage problem.
But there’s a good chance all these headlines are just a distraction. The real problem is datacentres using vast energy to train AI to better manipulate us. There is a big push to take the heat off them by focusing on bitcoin.
Yeah, people have been coming up with these ideas for at least a decade, but no actually effective CAPTCHA has ever been developed. Even if we did, rich people would just have buildings full of slaves doing CAPTCHAs =\
The whole premise of PoW is that
more money investment == more mining power, this is just as true for “memory constrained” algos as it is for the current hashing one. So the only thing changing the PoW would do is piss off investors by causing them to reinvest.I’m not saying It’s impossible for us to come up with a better system, its just that I haven’t seen any compelling replacements.
yes. so instead of competing for electricity, they would start competing for memory cards. so in the end they would use less electricity.
the second benefit is to change the puzzle to take away the pros’ advantage. make the requirements to play so that it’s easy for home users but not much easier for pros.
In general
more memory cards == more power, but even if this wasn’t the case, then the price of memory chips/ silicon would go up and it would be even harder to buy a video card =/ The people/ places that had better access to chips would wield power over the network. The nice thing about electricity is there are a bunch of ways to make it, so the playing field is pretty level.This would be nice, but none of the conceived ideas have been made to be robust. Sybil attacks are seemingly always possible in a decentralized network. There are some mitigation strategies but nothing bullet proof I think so far.
Bitcoins strategy was to rely on greedy investors to secure the network. I think of bitcoin more of as an experiment than anything, so changing it would ruin the experiment. A more socially-oriented network would have to supersede it from the outside for anything to change bitcoin realistically.
That’s interesting about Sybil attacks. I’ve thought about that - my idea was to have a global online convention, once per year, that lasts nine minutes. During that time you must pass a turing test from an existing user, and then give a turing test to two other (new or existing) users. So it would be very hard to obtain multiple identities, without being a real user.
The turing tests can be video calls where you ask each other to draw pictures, on video, of things that have never existed.
Actually that’s not mentioned in the wiki article. Maybe I’m the first person to think of it.
I’ve been thinking a bit about what you said. If the puzzle were changed, so that instead of just competing for electricity and GPUs, they are competing for RAM cards too … how would that affect the difficulty and the strain on global resources (not just talking about electricity now).
You change the mining function somehow, that makes it much much more expensive for pros, but are free for normal people mining on laptops. Some kind of function that requires:
allocating 1GB RAM and 10GB disk space
human input
once a block is made on this computer, it can’t start work on another block for a while
can only mine for 8 hours a day
That sort of thing. The result, I think, would be this:
More people will mine. The difficulty will go up. The share of blocks going to pros will go down. The profitability for pros will go down in proportion with the share of blocks they are getting. Many pros will stop mining. The global resources used for mining will go down.
EDIT: After not much research, somebody has thought of this before.
The example of a mining function like this (i had to look it up again) is Monero and randomX. Proof of allocating disk space also already exists. And there is also a currency which changes mining algorithm every block, but I forget which one.
It’s an idea a lot of people have been working on. If my logic above is sound, this approach not only solves the electricity wastage problem, but also the hardware wastage problem.
But there’s a good chance all these headlines are just a distraction. The real problem is datacentres using vast energy to train AI to better manipulate us. There is a big push to take the heat off them by focusing on bitcoin.