well I was merely flavoring, right?
The discussion was about how someone can relatively easily own > 50% in a PoS network.
In case of Proof of Work, hardware manufacturers are in that position
Last edited:
CryptoCurrency Mining with GPUs *spawn*
- Thread starter CarstenS
- Start date
The discussion was about how someone can relatively easily own > 50% in a PoS network.
In case of Proof of Work, hardware manufacturers are in that position
Jawed
Legend
Except that the other miners, if they detect an attack, will orphan blocks that are an attack. An attack would be something along the lines of creating extra coins or double-spending.
There's no reason for the other miners who witness an attack to accept it.
Honest miners who happen to have more than 50% hash rate aren't attacking, while they produce blocks.
Having more than 50% hash rate isn't an attack per se. When Satoshi started mining he was 100% of the hash rate.
Jawed
Legend
In bitcoin, the timings and counts of blocks required for "safety" were designed deliberately.
[Edited paragraph to make it clearer] So the 10 minute timing of blocks is a tuned value, it's not something plucked out of thin air. It relates to planetwide networking latencies and the probability distribution over time of newly mined blocks and takes account of the change over time (from day 1 to well into the future) of network usage. So you can get 2 blocks mined one second apart but you can also get two blocks mined 1 hour apart. Both of these are extremely unlikely with a 10 minute interval as seen in bitcoin, and the chances are lower as hashrate rises.
The generally accepted 6 blocks confirmation for "large value" transactions caters to the long term network behaviour. In the early days the variations in duration between blocks were large but they tend to reduce as hashrate increases.
The 100 block delay, for spendable coins produced in a mined block, is to allow the network to easily deal with certain kinds of rogue miner behaviour.
But note that proof of work ends with mining in bitcoin. Once you talk about the mechanics of smart contracts, then you are working with a new system that's laid on top of the underlying network (e.g. bitcoin).
This is similar to how the world wide web is a set of mechanics that runs on top of the TCP/IP protocols. HTTPS security for websites is a feature added on top.
So proof of stake in smart contracts is the cause of this kind of problem. It's entirely possible to put a proof of work mechanism into smart contracts and it's possible to program in certain kinds of throttles so that, for example, a newly attained majority share holder is forced to wait a day or a week before they can use their majority position in any way.
So borrowing $1B for 1 day or a week makes the attack uneconomic.
So what this particular attack shows is that people who design these crypto-economic systems are usually complete fuckwits who can't see beyond the end of their nose, rubbing their hands in glee at the profits they can earn with their cool new concept. I cheer along with everyone else when these idiots get hacked, because they deserve to get owned as do their customers.
Governance has been a red flag in these kinds of smart contract systems since day 1.
Last edited:
Because it increases the chance of a majority attack as in the above heist.
POW, by being competitive, favors the lowest cost solution.
Economies of scale guarantees big miners a big advantage who slowly concentrate into few monopolies. Besides being an unsightly waster of energy in a world with energy uncertainty while rolling towards climate change doom, it is also completely contrary to the original idea with decentralized transactions/crowd sourcing of monetary authority.
Cheers
Jawed
Legend
It doesn't. Honest miners will spot the attack and reject it. It's not rocket science. All of the data required to spot the attack is available to all miners.
All honest miners have a life-or-death incentive to ensure that the network continues to be seen as trustworthy, which requires that attacks by dishonest miners are rejected.
Yes. Cost is not static, so if you're winning today with your lower costs, you might not be winning tomorrow when someone else comes up with lower costs.
Coopetition - Wikipedia
There's no such thing as "few monopolies" in a single competitive arena.
Bitcoin mining was always intended to be performed by large-scale operations. November 2008, months before the original version of the software was released:
Bitcoin P2P e-cash paper | Satoshi Nakamoto Institute
Decentralisation in bitcoin is specifically about not being required to trust a single entity's version of the truth.
The miners, in competition with each other, are incentivised to protect the trustworthiness of the network, as a whole. Otherwise their investment in hardware (and staff to maintain it) has no return.
In the early days of bitcoin, hardware costs had a return on investment measured in days or weeks. It's not so pretty now
Is this story about some NFT sale by the Bored Ape Yacht Club destroying a ton of etherium true? If so could you please explain it to me like I'm old and can't understand how anyone thinks crypto and NFTs are any kind of "investment".
Jawed
Legend
Rootax
Veteran
I believe these ape/crap nft is the dumbest trend. But a project screwing up launch soooo the whole thing is fucked ? Their smart contract, minting, whatever, was bad from the start, especially given the strength and weakness of the eth blockchain. Somebody crashing on the highway and creating a huge roadblock doesn't mean highway are bad...
D
Deleted member 2197
Guest
Jawed
Legend
SEC.gov | SEC Charges NVIDIA Corporation with Inadequate Disclosures about Impact of Cryptomining
$5.5M fine pays for SEC staff time?
$5.5M fine pays for SEC staff time?
Jawed
Legend
Maybe some investors will have another go?:
Nvidia Wins Billion-Dollar Crypto Mining Lawsuit - ExtremeTech
Nvidia Wins Billion-Dollar Crypto Mining Lawsuit - ExtremeTech
Man from Atlantis
Veteran
100% LHR unlock at NiceHash? It’s here!
This won't work with 3080 12GB and 3050, which features the newest LHR unlock that hackers couldn't get.
This won't work with 3080 12GB and 3050, which features the newest LHR unlock that hackers couldn't get.
The one's from nbminer on hive os also works
Silent_Buddha
Legend
‘I lost my life savings’: Terra Luna cryptocurrency collapses 98% overnight (yahoo.com)
Ouch. I almost feel sorry for some of those people ... almost. But it's their own stupidity for sinking their life savings into an unregulated alt-currency that's more unstable than investing in stocks on the stock market.
At least it recovered a tiny bit...
Terra’s Luna cryptocurrency is down 87% in one day, but at least it’s not down 95% anymore. Its sister stablecoin was at 29 cents when it should have been $1 (yahoo.com)
Regards,
SB
Ouch. I almost feel sorry for some of those people ... almost. But it's their own stupidity for sinking their life savings into an unregulated alt-currency that's more unstable than investing in stocks on the stock market.
At least it recovered a tiny bit...
Terra’s Luna cryptocurrency is down 87% in one day, but at least it’s not down 95% anymore. Its sister stablecoin was at 29 cents when it should have been $1 (yahoo.com)
Regards,
SB
seems to be the same issue with NFT.
btw I'm still not sure what blockchain is great for. the awesome things that people says blockchain is great for (e.g. transferring items from game A to B, automatic fees cut for NFT artists, etc) can be done with conventional database (and maybe even faster, less power consumption, etc)
Blockchain as a distributed ledger could seriously speed up transactions in the financial industry, eliminate a lot of errors and remove the need for trade matching and reconciliation. But it would require all the banks, brokers, exchanges, hedge funds etc to agree on a single shared network and protocol. Never going to happen.
Similar threads
