3 ms·

Forgive my stupidity, but I don't understand this statement by Ben (Stratechery): "In practice, it is much more complicated: while a limited set of “validators

Forgive my stupidity, but I don't understand this statement by Ben (Stratechery): "In practice, it is much more complicated: while a limited set of “validators” — aka miners — share a history of transactions in (individual) blocks that are chained together (i.e. a blockchain), what Libra actually exposes is the current state of the ledger. In practice this means that adding new transactions can be much quicker and more efficient — more akin to adding a line to a spreadsheet than rebuilding the entire spreadsheet from scratch." How is this different than Bitcoin, and why does it require a different level of trust? You are still trusting the validators (miners) either way, correct? And how is it more efficient?

snarf21 7y ago
Libra intends to use Proof of Stake. This means that the initial members will "stake" some amount of Libra and now all transactions can only be validated by them. You as an individual can't validate any new transaction. The assumption is that because the PoS validator has a huge stake in this currency, they will behave rationally or otherwise they would devalue their own stake. The trust here is completely centralized and is a rich get richer model. If FB or PayPal or a member doesn't want to add your transaction, they can completely block you. They control what is true. Bitcoin uses Proof of Work. Consensus is decentralized. Anyone can do some "work" (hashing in this case). If they find the magic number to get the right hash, they get rewarded by the Bitcoin protocol itself. This is also the only way new Bitcoins are created. It is only when one of these magic numbers gets found that transactions get added to the Bitcoin ledger (every 10 minutes). The rule is that the longest ledger chain wins so everyone is running in parallel. You don't need to own any Bitcoin to do mining to get "free" Bitcoin. The trust is decentralized (in terms of control) and no one can prevent anyone else from validating Bitcoin transactions as long as you follow the same rules as everyone else.
- creeble 7y ago
  Probably too late to keep this discussion going, but... I understand Proof of Stake vs. Proof of Work. But what Ben says about only "publishing the current state of the ledger" doesn't seem to relate to whether PoS is used. It's certainly not inherent to PoS that the entire blockchain isn't published. Maybe he's just confused about how PoS works (but I somehow doubt that). I understand how PoS makes things more efficient by entrusting a single validator (or any small subset of validators) rather than having essentially a contest, and I understand why this is also inherently less trustworthy than PoW. But Ben never mentions Proof of Stake anywhere in his explanation, just that the blockchain's "current state" is the only thing "exposed". I'm juse trying to understand what he's getting at there.
  - snarf21 7y ago
    Gotcha, I assumed that it was a little handwavy on the details to be more accessible to a wider audience. It is very possible that they don't let others download the complete ledger and only let you ask for a certain private key's current balance. I agree that this is even more concerning (if true) than regular PoS. I also didn't read the proposal deep enough to know if you can register to be a node that sees all proposed transactions even though you can't actually validate them. If there is on way for external monitoring or audit, this truly is not blockchain even in name only. Then it is simply a replicated database with restricted access.
danans 7y ago
> You are still trusting the validators (miners) either way, correct? And how is it more efficient? Who can be a validator (miner) is controlled by the consortium, and therefore the blockchain algorithm doesn't need be to as computationally expensive as bitcoin's proof of work, which is designed to contend with the fact that anyone can be a miner. Instead it can just serve the purpose of enabling a distributed ledger.

snarf21 7y ago

Libra intends to use Proof of Stake. This means that the initial members will "stake" some amount of Libra and now all transactions can only be validated by them. You as an individual can't validate any new transaction. The assumption is that because the PoS validator has a huge stake in this currency, they will behave rationally or otherwise they would devalue their own stake. The trust here is completely centralized and is a rich get richer model. If FB or PayPal or a member doesn't want to add your transaction, they can completely block you. They control what is true. Bitcoin uses Proof of Work. Consensus is decentralized. Anyone can do some "work" (hashing in this case). If they find the magic number to get the right hash, they get rewarded by the Bitcoin protocol itself. This is also the only way new Bitcoins are created. It is only when one of these magic numbers gets found that transactions get added to the Bitcoin ledger (every 10 minutes). The rule is that the longest ledger chain wins so everyone is running in parallel. You don't need to own any Bitcoin to do mining to get "free" Bitcoin. The trust is decentralized (in terms of control) and no one can prevent anyone else from validating Bitcoin transactions as long as you follow the same rules as everyone else.

creeble 7y ago
Probably too late to keep this discussion going, but... I understand Proof of Stake vs. Proof of Work. But what Ben says about only "publishing the current state of the ledger" doesn't seem to relate to whether PoS is used. It's certainly not inherent to PoS that the entire blockchain isn't published. Maybe he's just confused about how PoS works (but I somehow doubt that). I understand how PoS makes things more efficient by entrusting a single validator (or any small subset of validators) rather than having essentially a contest, and I understand why this is also inherently less trustworthy than PoW. But Ben never mentions Proof of Stake anywhere in his explanation, just that the blockchain's "current state" is the only thing "exposed". I'm juse trying to understand what he's getting at there.
- snarf21 7y ago
  Gotcha, I assumed that it was a little handwavy on the details to be more accessible to a wider audience. It is very possible that they don't let others download the complete ledger and only let you ask for a certain private key's current balance. I agree that this is even more concerning (if true) than regular PoS. I also didn't read the proposal deep enough to know if you can register to be a node that sees all proposed transactions even though you can't actually validate them. If there is on way for external monitoring or audit, this truly is not blockchain even in name only. Then it is simply a replicated database with restricted access.

danans 7y ago

> You are still trusting the validators (miners) either way, correct? And how is it more efficient? Who can be a validator (miner) is controlled by the consortium, and therefore the blockchain algorithm doesn't need be to as computationally expensive as bitcoin's proof of work, which is designed to contend with the fact that anyone can be a miner. Instead it can just serve the purpose of enabling a distributed ledger.