4 ms·

I'm not sure what you mean when you say that Bitcoin can scale by using 3rd party payment providers? Any Bitcoin transaction will only ever be as fast as the un

I'm not sure what you mean when you say that Bitcoin can scale by using 3rd party payment providers? Any Bitcoin transaction will only ever be as fast as the underlying blockchain, which currently processes around seven transactions per second (last time I checked). Of course proof of stake can be decentralized! I think what you mean is that it's not clear whether a widely adopted PoS based blockchain will end up centralizing power in the hands of a few whales. This is a legitimate concern; however, this threat is no different than the threat of centralization of computing power in Bitcoin. In Bitcoin, this isn't even a threat, it’s a reality! I'll say it again: Bitcoin and Ethereum are not great currencies. There are altcoins that have far surpassed both of them in terms of achieving security, scalability, and decentralization simultaneously.

wanderingmind 5y ago
Can you provide more details/references to your statement on altcoins that have better scalability and decentralization.
- riptheworld 5y ago
  [dead]
  - javert 5y ago
    > I don’t want to turn this thread into a crypto version of Wall Street Bets where people are arguing over their favorite coins it seems like that's exactly what you want. Given that you are spreading lots of misinformation and don't seem to be trying to understand the other side's point.
  - tromp 5y ago
    > The Algorand cryptocurrency achieves thousands of transactions per second And is it possible to sync to the current Algorand state by downloading and verifying the complete transaction history, like I can with Bitcoin? And if so, how big is that download? Bitcoin limits the tx rate on purpose so that the Initial Block Download remains feasible on average hardware, and no trust is needed beyond the integrity of the client software.
    - riptheworld 5y ago
      Bitcoin’s throughput is limited as a function of network security and network delay. It has nothing to do with making it easier for clients to sync with the current state. You can improve Bitcoin’s transaction rate in two ways: 1.) decrease mining difficulty, i.e., the amount of hashing required to produce a valid proof-of-work on average. This impacts the security of the network. 2.) increase the block size so you can fit more transactions into a single block. This negatively impacts network latency (and therefore scalability, as it increases the burden of propagating messages across the network). To your question on Algorand: yes, new consensus participants can quickly and easily sync with the current state of the blockchain through fast catch-up [1], without any meaningful security degradation [2]. [1] https://youtu.be/BOnond-J3Zo https://youtu.be/BOnond-J3Zo [2] https://eprint.iacr.org/2018/269.pdf https://eprint.iacr.org/2018/269.pdf
      - tromp 5y ago
        So there's no way to trustlessly verify Algorand history; you have to trust a checkpoint, or trust some signatory quorum in some alternative design that's not even implemented yet.
        riptheworld 5y ago
        You can trustlessly verify the entire Algorand history by verifying it yourself (same as Bitcoin). However, almost all blockchains (including Bitcoin) will suffer from this problem of longer and longer sync times, and so some bootstrapping solution is necessary.
kayamon 5y ago
> Any Bitcoin transaction will only ever be as fast as the underlying blockchain, which currently processes around seven transactions per second (last time I checked). Then check again. Go google the Lightning Network. This is outdated FUD. Lightning sets up secure channels on top of the underlying blockchain. Actual blockchain transactions are only required (I'm simplifying) to record slow trends in average usages.
- riptheworld 5y ago
  You’re confusing a layer 2 solution that provides separate (and weakened) transaction assurances than the Bitcoin protocol itself. This is evident in the fact that the lightning network suffers from security vulnerabilities (e.g., wormhole attack [1]) not present in the Layer 1 Bitcoin protocol. This is not to say that the Lightning Network doesn't significantly improve the usability of Bitcoin as a currency. But it does so at the expense of security. A much better approach is to use a layer 1 solution that achieves scalability, security, and decentralization by default. [1]https://eprint.iacr.org/2018/472.pdf https://eprint.iacr.org/2018/472.pdf
javert 5y ago
> I'm not sure what you mean when you say that Bitcoin can scale by using 3rd party payment providers? Try thinking about it. It would work the same way PayPal works with USD. Bitcoin payments simply don't need to happen on the blockchain. Similarly, I can buy gold in my brokerage account without any actual gold moving between warehouses. > Of course proof of stake can be decentralized! I think what you mean is that it's not clear whether a widely adopted PoS based blockchain will end up centralizing power in the hands of a few whales. This is a legitimate concern That's right, which is why it probably can't be decentralized. Even if the majority of holders are small individuals, the coins still end up on exchanges, which become whales. > however, this threat is no different than the threat of centralization of computing power in Bitcoin. This is a common misunderstanding of Bitcoin. Bitcoin miners do not "vote" or participate in governance. Thus, bitcoin does not have vulnerabilities that proof of stake has. Bitcoin's decentralization does not come from miners. It comes from users who run nodes and enforce the rules of bitcoin by not recognizing illegitimate protocol changes, invalid blocks, etc. Market forces also play an important part: Bitcoiners would not pay for coins from a block that breaks the rules of bitcoin, because those coins will not be recognized as being bitcoin.
- riptheworld 5y ago
  Oh sheesh. No one said anything about Bitcoin miners voting. I was simply pointing out that centralization of mining power (which has already happened in Bitcoin) has essentially the same security impact as centralization of wealth in proof of stake consensus mechanisms. > It would work the same way PayPal works with USD. Bitcoin payments simply don't need to happen on the blockchain. Similarly, I can buy gold in my brokerage account without any actual gold moving between warehouses. This is a terrible idea.
  - javert 5y ago

wanderingmind 5y ago

Can you provide more details/references to your statement on altcoins that have better scalability and decentralization.

riptheworld 5y ago
[dead]
- javert 5y ago
  > I don’t want to turn this thread into a crypto version of Wall Street Bets where people are arguing over their favorite coins it seems like that's exactly what you want. Given that you are spreading lots of misinformation and don't seem to be trying to understand the other side's point.
- tromp 5y ago
  > The Algorand cryptocurrency achieves thousands of transactions per second And is it possible to sync to the current Algorand state by downloading and verifying the complete transaction history, like I can with Bitcoin? And if so, how big is that download? Bitcoin limits the tx rate on purpose so that the Initial Block Download remains feasible on average hardware, and no trust is needed beyond the integrity of the client software.
  - riptheworld 5y ago
    Bitcoin’s throughput is limited as a function of network security and network delay. It has nothing to do with making it easier for clients to sync with the current state. You can improve Bitcoin’s transaction rate in two ways: 1.) decrease mining difficulty, i.e., the amount of hashing required to produce a valid proof-of-work on average. This impacts the security of the network. 2.) increase the block size so you can fit more transactions into a single block. This negatively impacts network latency (and therefore scalability, as it increases the burden of propagating messages across the network). To your question on Algorand: yes, new consensus participants can quickly and easily sync with the current state of the blockchain through fast catch-up [1], without any meaningful security degradation [2]. [1] https://youtu.be/BOnond-J3Zo https://youtu.be/BOnond-J3Zo [2] https://eprint.iacr.org/2018/269.pdf https://eprint.iacr.org/2018/269.pdf
    - tromp 5y ago
      So there's no way to trustlessly verify Algorand history; you have to trust a checkpoint, or trust some signatory quorum in some alternative design that's not even implemented yet.
      - riptheworld 5y ago
        You can trustlessly verify the entire Algorand history by verifying it yourself (same as Bitcoin). However, almost all blockchains (including Bitcoin) will suffer from this problem of longer and longer sync times, and so some bootstrapping solution is necessary.

kayamon 5y ago

> Any Bitcoin transaction will only ever be as fast as the underlying blockchain, which currently processes around seven transactions per second (last time I checked). Then check again. Go google the Lightning Network. This is outdated FUD. Lightning sets up secure channels on top of the underlying blockchain. Actual blockchain transactions are only required (I'm simplifying) to record slow trends in average usages.

riptheworld 5y ago
You’re confusing a layer 2 solution that provides separate (and weakened) transaction assurances than the Bitcoin protocol itself. This is evident in the fact that the lightning network suffers from security vulnerabilities (e.g., wormhole attack [1]) not present in the Layer 1 Bitcoin protocol. This is not to say that the Lightning Network doesn't significantly improve the usability of Bitcoin as a currency. But it does so at the expense of security. A much better approach is to use a layer 1 solution that achieves scalability, security, and decentralization by default. [1]https://eprint.iacr.org/2018/472.pdf https://eprint.iacr.org/2018/472.pdf

javert 5y ago

> I'm not sure what you mean when you say that Bitcoin can scale by using 3rd party payment providers? Try thinking about it. It would work the same way PayPal works with USD. Bitcoin payments simply don't need to happen on the blockchain. Similarly, I can buy gold in my brokerage account without any actual gold moving between warehouses. > Of course proof of stake can be decentralized! I think what you mean is that it's not clear whether a widely adopted PoS based blockchain will end up centralizing power in the hands of a few whales. This is a legitimate concern That's right, which is why it probably can't be decentralized. Even if the majority of holders are small individuals, the coins still end up on exchanges, which become whales. > however, this threat is no different than the threat of centralization of computing power in Bitcoin. This is a common misunderstanding of Bitcoin. Bitcoin miners do not "vote" or participate in governance. Thus, bitcoin does not have vulnerabilities that proof of stake has. Bitcoin's decentralization does not come from miners. It comes from users who run nodes and enforce the rules of bitcoin by not recognizing illegitimate protocol changes, invalid blocks, etc. Market forces also play an important part: Bitcoiners would not pay for coins from a block that breaks the rules of bitcoin, because those coins will not be recognized as being bitcoin.

riptheworld 5y ago
Oh sheesh. No one said anything about Bitcoin miners voting. I was simply pointing out that centralization of mining power (which has already happened in Bitcoin) has essentially the same security impact as centralization of wealth in proof of stake consensus mechanisms. > It would work the same way PayPal works with USD. Bitcoin payments simply don't need to happen on the blockchain. Similarly, I can buy gold in my brokerage account without any actual gold moving between warehouses. This is a terrible idea.
- javert 5y ago