6 ms·

This just isn't true. Yes, running your own blockchain node is infeasible, but a blockchain-based service with hosted nodes is better than a federated service.

by natrius 10y ago

This just isn't true. Yes, running your own blockchain node is infeasible, but a blockchain-based service with hosted nodes is better than a federated service. Proof-of-work creates an illusion of dependence on blockchain miners. Proof-of-stake makes it clear that blockchains are run by a corporation that wants to maximize transaction fee revenue. Transaction validators have to do the best job of maintaining a unified historical record of all transactions for people to keep giving them business. Otherwise, users will choose different validators and carry on without them. Blockchains are fancy spools of receipt paper. If the printer gets jammed, tear off the stuff you care about and staple it to a new spool. The printer doesn't control the people. The people employ the printer. With that said, it's still unclear whether blockchains are the right choice for publishing. A decentralized Twitter doesn't need consensus on a historical record. An IPFS-based Twitter is probably a better way to start, then anchor the parts that need consensus on a blockchain, like usernames and timestamps.

api 10y ago
Your last point is important-- I think too many people in the distributed systems / decentralization community rush to block chain as the one hammer to pound every nail. It's not. It's a solution to a very specific set of consensus problems but outside that it's not the best thing.
goldbrick 10y ago
What part isn't true, specifically? We are dependent on blockchain miners, because without them our transactions don't get accepted. Moreover we have a couple established big players that control the market and no clear way for them to get disrupted, save abandoning Bitcoin entirely. Blockchains are much more than receipt paper. Receipt paper is a superfluous byproduct. The blockchain is the system of record.
- natrius 10y ago
  This is untrue: "Because blockchain is consensual, after a certain point of centralization, the rules of the system depend on very few users." If a blockchain has rules that don't meet my needs, I can choose a different blockchain. If I adopt a blockchain that changes its rules to be undesirable, I can take the history I care about and move it to another blockchain. This argument depends on proof-of-stake. Some people think proof-of-work blockchains are the only useful ones, but I disagree. It's also a boring debate to have, so I won't be doing that here.
  - drdaeman 10y ago
    > If a blockchain has rules that don't meet my needs, I can choose a different blockchain. Had this happened with Bitcoin? I think I heard there are some quite major disagreements (something about block sizes and stuff) and there are a lot of users that aren't happy. I don't remember hearing them just "well, whatever" and forking off.
    - natrius 10y ago
      Bitcoin is a special case because it's only really intended to be used as a ledger for bitcoins. Currencies are a collective delusion: they're valuable because other people think they are valuable. Any dispute over what the source of truth is for a currency's ledger strains that delusion, possibly to the breaking point. The value of the currency would suffer, which would defeat the purpose of a fork. Ethereum is a general purpose blockchain. Forking it could hurt the value of ether, but for many applications, the value of ether doesn't matter. Moving a blockchain-based Twitter from one fork to another is relatively easy and consequence-free.
      - drdaeman 10y ago
        > Moving a blockchain-based Twitter from one fork to another is relatively easy and consequence-free. I'm not sure it's really consequence-free. It's essentially a netsplit, where the forking party lose connectivity with everyone who's left. Consider the case where Twitter (the centralized one) is suddenly falling apart over some issue into two distinct non-interoperating services. Technically - sure - it's a no-brainer, but socially it feels quite complicated to me. (I guess, this would work if the app would be designed to use multiple blockchains. I need to give it some thought.)
        natrius 10y ago
        You got it. Push a new client that supports both chains to users with enough time before the switch for people to upgrade.
- jon_richards 10y ago
  One idea I had was if it is possible to somehow have all the miners "guessing" where the next block is going to be signed. A correct guess would net a bit of the transaction fee. Huge server farms concentrated in an area would therefore lose out, because everyone would guess that location. The biggest difficulty would be accurately identifying where the signature came from, but with clever timing mechanisms, it might be possible given the latency of light. I haven't thought about it much, but maybe require the current time be included in what gets hashed. Anyone cheating by using a future time could be caught by looking for missing transactions in that time space.
  - erikpukinskis 10y ago
    I believe Ethereum does something like this: "The Casper protocol is intended to offer stronger finality guarantees than proof of work. First, there is a standard definition of “total economic finality”: it takes place when 2/3 of all validators make maximum-odds bets that a given block or state will be finalized. This condition offers very strong incentives for validators to never try to collude to revert the block: once validators make such maximum-odds bets, in any blockchain where that block or state is not present, the validators lose their entire deposits. As Vlad Zamfir put it, imagine a version of proof of work where if you participate in a 51% attack your mining hardware burns down." https://blog.ethereum.org/2016/05/09/on-settlement-finality/ https://blog.ethereum.org/2016/05/09/on-settlement-finality/
schmichael 10y ago
> Proof-of-work creates an illusion of dependence on blockchain miners. How is proof-of-work's dependence on miners an illusion? Seems like a fundamental property of proof-of-work that it depends solely on the consensus of miners.
- natrius 10y ago
  Miners only mine to gain more bitcoins. The market determines what bitcoins are. The market can decide to treat a different historical record with different rules as the ledger for bitcoins. This is much easier in proof-of-stake systems. You can't destroy a bad miner's mining equipment, but you can destroy a bad validator's deposit, which removes them from the system.
  - falcolas 10y ago
    > you can destroy a bad validator's deposit And how does this happen? By a consensus of miners? If I, as an individual, were to find reason to distrust a entity who makes up the majority of mining transactions, how likely is it that I will be able to use the main blockchain ever again?
    - natrius 10y ago
      Yes, by a consensus of miners, or by users choosing a new set of miners. The "main blockchain" is not a decision of the miners. It's a decision of the people. If everyone agrees that current miners are behaving badly in a proof-of-stake system, they can just choose new miners.
      - toss1 10y ago
        OK, I'll bite, with a serious, practical question -- how would I do that this week? Please assume that I have 100 BTC at a particular address or small set of addresses, and I don't like the Chinese mining consortia. How do I actually move my value to a new chain that 1) excludes those mining consortia, 2) preserves my stored value, and 3) preserves my ability to transact with any other valid BTC address? How can I get this done by next week?
athenot 10y ago
> but a blockchain-based service with hosted nodes is better than a federated service ...and a federated service is better than a handful of silos controlling the network. So if the blockchain is supported by the mining of many small miners, it's a decentralized dream. But if there are a handful of miners validating the majority of transactions, then it ends up smelling a lot like the handful of silos situation. Not exactly the same but very similar.
- natrius 10y ago
  Miners cannot control your eyes. You can choose any record as your source of truth. If miners produce a good one, look at that one. If they produce a bad one, look somewhere else.
  - t1amat 10y ago
    This sounds great but it is a dramatic misunderstanding of the real problem. Sure, you can personally decide you don't like a particular fork. However, the value of your position is not determined by you, it is determined by the community and their overall choice. Bitcoin forking is dangerous. If it goes bad and the community becomes overly fragmented then everyone loses. The strength of a position in Bitcoin is a combination of many factors, but the most important ones are related to the size of the community, its relative stability as a crypto coin, and the relative ubiquity of it.
RyJones 10y ago
Intel recently unveiled PoET - proof of elapsed time - in Sawtooth Lake. https://intelledger.github.io/introduction.html https://intelledger.github.io/introduction.html
mr_sturd 10y ago
> An IPFS-based Twitter is probably a better way to start Did you see the AKASHA discussion, last week? https://news.ycombinator.com/item?id=11678810 https://news.ycombinator.com/item?id=11678810
wslh 10y ago
> Proof-of-work creates an illusion of dependence on blockchain miners. PoW depends on blockchain miners by definition and the Bitcoin experience shows that at the end you have a concentration of power. > With that said, it's still unclear whether blockchains are the right choice for publishing. It is very clear that blockchains ARE NOT the right choice for publishing, its limitations for that purpose are based on strong computer science principles on distributed systems, consensus and network speed.
- natrius 10y ago
  Technological frontiers are no place for dogma. Sure, blockchains probably aren't a good choice for publishing, but a culture that constrains people's technology choices is incompatible with progress.
  - wslh 10y ago
    This is not dogma but computer science research, you need to find the right tools for the right problem. BTW one of my companies is focused in blockchain, cryptocurrencies, etc and I can't believe the level of hype experienced there. For example, I read about the DAO Ethereum crowdsale but we experience serious technical issues with Ethereum that clash with that hype.

by natrius 10y ago

api 10y ago

Your last point is important-- I think too many people in the distributed systems / decentralization community rush to block chain as the one hammer to pound every nail. It's not. It's a solution to a very specific set of consensus problems but outside that it's not the best thing.

goldbrick 10y ago

What part isn't true, specifically? We are dependent on blockchain miners, because without them our transactions don't get accepted. Moreover we have a couple established big players that control the market and no clear way for them to get disrupted, save abandoning Bitcoin entirely. Blockchains are much more than receipt paper. Receipt paper is a superfluous byproduct. The blockchain is the system of record.

natrius 10y ago
This is untrue: "Because blockchain is consensual, after a certain point of centralization, the rules of the system depend on very few users." If a blockchain has rules that don't meet my needs, I can choose a different blockchain. If I adopt a blockchain that changes its rules to be undesirable, I can take the history I care about and move it to another blockchain. This argument depends on proof-of-stake. Some people think proof-of-work blockchains are the only useful ones, but I disagree. It's also a boring debate to have, so I won't be doing that here.
- drdaeman 10y ago
  > If a blockchain has rules that don't meet my needs, I can choose a different blockchain. Had this happened with Bitcoin? I think I heard there are some quite major disagreements (something about block sizes and stuff) and there are a lot of users that aren't happy. I don't remember hearing them just "well, whatever" and forking off.
  - natrius 10y ago
    Bitcoin is a special case because it's only really intended to be used as a ledger for bitcoins. Currencies are a collective delusion: they're valuable because other people think they are valuable. Any dispute over what the source of truth is for a currency's ledger strains that delusion, possibly to the breaking point. The value of the currency would suffer, which would defeat the purpose of a fork. Ethereum is a general purpose blockchain. Forking it could hurt the value of ether, but for many applications, the value of ether doesn't matter. Moving a blockchain-based Twitter from one fork to another is relatively easy and consequence-free.
    - drdaeman 10y ago
      > Moving a blockchain-based Twitter from one fork to another is relatively easy and consequence-free. I'm not sure it's really consequence-free. It's essentially a netsplit, where the forking party lose connectivity with everyone who's left. Consider the case where Twitter (the centralized one) is suddenly falling apart over some issue into two distinct non-interoperating services. Technically - sure - it's a no-brainer, but socially it feels quite complicated to me. (I guess, this would work if the app would be designed to use multiple blockchains. I need to give it some thought.)
      - natrius 10y ago
        You got it. Push a new client that supports both chains to users with enough time before the switch for people to upgrade.
jon_richards 10y ago
One idea I had was if it is possible to somehow have all the miners "guessing" where the next block is going to be signed. A correct guess would net a bit of the transaction fee. Huge server farms concentrated in an area would therefore lose out, because everyone would guess that location. The biggest difficulty would be accurately identifying where the signature came from, but with clever timing mechanisms, it might be possible given the latency of light. I haven't thought about it much, but maybe require the current time be included in what gets hashed. Anyone cheating by using a future time could be caught by looking for missing transactions in that time space.
- erikpukinskis 10y ago
  I believe Ethereum does something like this: "The Casper protocol is intended to offer stronger finality guarantees than proof of work. First, there is a standard definition of “total economic finality”: it takes place when 2/3 of all validators make maximum-odds bets that a given block or state will be finalized. This condition offers very strong incentives for validators to never try to collude to revert the block: once validators make such maximum-odds bets, in any blockchain where that block or state is not present, the validators lose their entire deposits. As Vlad Zamfir put it, imagine a version of proof of work where if you participate in a 51% attack your mining hardware burns down." https://blog.ethereum.org/2016/05/09/on-settlement-finality/ https://blog.ethereum.org/2016/05/09/on-settlement-finality/

schmichael 10y ago

> Proof-of-work creates an illusion of dependence on blockchain miners. How is proof-of-work's dependence on miners an illusion? Seems like a fundamental property of proof-of-work that it depends solely on the consensus of miners.

natrius 10y ago
Miners only mine to gain more bitcoins. The market determines what bitcoins are. The market can decide to treat a different historical record with different rules as the ledger for bitcoins. This is much easier in proof-of-stake systems. You can't destroy a bad miner's mining equipment, but you can destroy a bad validator's deposit, which removes them from the system.
- falcolas 10y ago
  > you can destroy a bad validator's deposit And how does this happen? By a consensus of miners? If I, as an individual, were to find reason to distrust a entity who makes up the majority of mining transactions, how likely is it that I will be able to use the main blockchain ever again?
  - natrius 10y ago
    Yes, by a consensus of miners, or by users choosing a new set of miners. The "main blockchain" is not a decision of the miners. It's a decision of the people. If everyone agrees that current miners are behaving badly in a proof-of-stake system, they can just choose new miners.
    - toss1 10y ago
      OK, I'll bite, with a serious, practical question -- how would I do that this week? Please assume that I have 100 BTC at a particular address or small set of addresses, and I don't like the Chinese mining consortia. How do I actually move my value to a new chain that 1) excludes those mining consortia, 2) preserves my stored value, and 3) preserves my ability to transact with any other valid BTC address? How can I get this done by next week?

athenot 10y ago

> but a blockchain-based service with hosted nodes is better than a federated service ...and a federated service is better than a handful of silos controlling the network. So if the blockchain is supported by the mining of many small miners, it's a decentralized dream. But if there are a handful of miners validating the majority of transactions, then it ends up smelling a lot like the handful of silos situation. Not exactly the same but very similar.

natrius 10y ago
Miners cannot control your eyes. You can choose any record as your source of truth. If miners produce a good one, look at that one. If they produce a bad one, look somewhere else.
- t1amat 10y ago
  This sounds great but it is a dramatic misunderstanding of the real problem. Sure, you can personally decide you don't like a particular fork. However, the value of your position is not determined by you, it is determined by the community and their overall choice. Bitcoin forking is dangerous. If it goes bad and the community becomes overly fragmented then everyone loses. The strength of a position in Bitcoin is a combination of many factors, but the most important ones are related to the size of the community, its relative stability as a crypto coin, and the relative ubiquity of it.

RyJones 10y ago

Intel recently unveiled PoET - proof of elapsed time - in Sawtooth Lake. https://intelledger.github.io/introduction.html https://intelledger.github.io/introduction.html

mr_sturd 10y ago

> An IPFS-based Twitter is probably a better way to start Did you see the AKASHA discussion, last week? https://news.ycombinator.com/item?id=11678810 https://news.ycombinator.com/item?id=11678810

wslh 10y ago

> Proof-of-work creates an illusion of dependence on blockchain miners. PoW depends on blockchain miners by definition and the Bitcoin experience shows that at the end you have a concentration of power. > With that said, it's still unclear whether blockchains are the right choice for publishing. It is very clear that blockchains ARE NOT the right choice for publishing, its limitations for that purpose are based on strong computer science principles on distributed systems, consensus and network speed.

natrius 10y ago
Technological frontiers are no place for dogma. Sure, blockchains probably aren't a good choice for publishing, but a culture that constrains people's technology choices is incompatible with progress.
- wslh 10y ago
  This is not dogma but computer science research, you need to find the right tools for the right problem. BTW one of my companies is focused in blockchain, cryptocurrencies, etc and I can't believe the level of hype experienced there. For example, I read about the DAO Ethereum crowdsale but we experience serious technical issues with Ethereum that clash with that hype.