8 ms·

The article is pretty light on technical details. Does Ethereum suffer from the same design limitation as Bitcoin, where every operation must be verified by eve

The article is pretty light on technical details. Does Ethereum suffer from the same design limitation as Bitcoin, where every operation must be verified by every node? If so, I have a really hard time believing it will scale. Look how much difficulty the Bitcoin network is having with just a couple of transactions per second.

empath75 11y ago
That's a political problem with the miners and developers. There's no technical reason they couldn't raise the block size.
- ufo 11y ago
  You shouldn't ignore political problems when it comes to distributed systems... Additionally, raising the block size cap is only a stopgap measure. Eventually you will reach hard bandwidth or disk storage limits and the only way will be to throw some transactions outside the blockchain.
  - nickik 11y ago
    Bishares is a blockchain that can do as many and more transaction as Visa. Their does not seem to be a fundamental problem as long as you accept that Worker(miners) are fast servers with good networks.
- vonklaus 11y ago
  Is this true? I believe there is a tradeoff but want a more in depth explsnation if possible
  - maxander 11y ago
    As I recall (and this may not be the whole of it), the "tradeoff" was that a larger block size would risk getting attention from China's great firewall, effectively splitting the network and rendering Chinese users' funds effectively worthless. And it turns out a set of Chinese users have a controlling interest in the network currently... so yes, its a political problem par excellence. It occurs to me as I write that Ethereum may have similar problems down the line, when platform updates turn out to be direly needed, but would disrupt the operation of well-established and lucrative apps.
    - empath75 11y ago
      Also, the miners get more fees when the transaction limit gets reached.
Hermel 11y ago
The Bitcoin network is not having much difficulty with a couple of transactions per second. It is the community that struggles with adjusting to an appropriate throughput.
- vertex-four 11y ago
  How do I continue using Bitcoin as a transactional currency if I can't perform transactions with it?
  - vonklaus 11y ago
    well, you couldn't. Peter Todd is a btc dev and provided a lot of thinking on the tradeoffs. He stated in an interview (fairly confident in this) a 100x improvement in speed and is in favour of not increasing block size. Should check it out, the intetview was on a podcast on the LTB network on itunes. So you will be able to. He has done a lot of thinking around tree chains and scaling and optimiztion. Btc is limited inamount of coins and came first, its value is as a currency. Eth is the first popular blockchaindb. Coins are needed to pay miners and powet dapps. If a currency cant be traded it idnt worth anything, however sidechains seem to provide a way to convert from one to another
  - tommorris 11y ago
    I believe the current answer to that is to make vague promises about how the vapourware known as the Lightning Network will solve all the problems.
- Crespyl 11y ago
  "a couple" transactions per second is not enough for a global network that intends to have millions of participants.
jnpn 11y ago
The Ethereum foundation has been putting a lot of resources into scaling research. Theres some promising proposals involving sharding the validation process, in addition to the off-chain scaling protocols that Bitcoin is pursuing like the lightning network.
pjc50 11y ago
Isn't that inherent to blockchains (every node must have a copy of all transactions)?
- jsprogrammer 11y ago
  You don't need to keep all the transactions. The original paper addresses how to compact the chain. Basically, you only need to keep hashes.
heliumcraft 11y ago
It's not meant to be used for 'distributed' computation.
- eternalban 11y ago
  Any system based on distributed consensus is by definition a distributed system.
  - heliumcraft 11y ago
    right. But it's meant to be used for consensus and not heavy computations.
    - eternalban 11y ago
      The consensus bit is the "heavy computation". The means regardless of the ends.
davidgerard 11y ago
Every computation runs on every instance of Ethereum, yes. It is important to note that this is not a resource for distributed general computation - it's for running smart contracts. Per the Ethereum development tutorial, "you will not be able to do anything on the EVM that you cannot do on a smartphone from 1999." https://github.com/ethereum/wiki/wiki/Ethereum-Development-Tutorial https://github.com/ethereum/wiki/wiki/Ethereum-Development-T... This is not a very good article - it's rather breathless and short on technical detail. In particular, calling it "the World Computer" is completely not what Ethereum will be capable of.
- vonklaus 11y ago
  Ethereum refers to itself as this on the website, ethereum.com
crazydoggers 11y ago
One key difference with Bitcoin is that, unlike Bitcoin, Ethereum has an account based architecture. In other words, you have an account that is credited or debited, and you then have a current balance for that account. If you need to get your balance or send ether, you just look up a single number. Bitcoin on the other hand uses an "Unspent Transaction Output" architecture. Basically your balance is made up of all of the credits and debits (inputs and outputs) you've made with that account. So sending bitcoins requires referencing "unspent outputs".. which may be lots and lots of different transactions. That's one difference, and there are others, such as the time taken to confirm blocks (current block time on ethereum is 16s). But there are still lots of scalability issues to be addressed, which are actively being worked on. Vitalik Buterin has a great couple of posts on some of the issues: https://blog.ethereum.org/2014/09/17/scalability-part-1-building-top/ https://blog.ethereum.org/2014/09/17/scalability-part-1-buil... edit: Not sure why this was getting downvoted. Just to clarify, in case it seemed like I wasn't answering the parent. The points above are one of the ways in which "every operation must be verified by every node" is less costly on Ethereum. Obviously if it takes less compute time per transaction, you can scale to larger numbers of transactions per second. Ethereum has already passed Bitcoin in this regard, and has hit at least 8 transactions per seconds at one point, while Bitcoin's ceiling is 7 transactions per second. In addition, prior to launch I believe they managed to push the test network up to 25 transactions per second.
- Anonobread 11y ago
  But Ethereum's raison d'etre is "X for the blockchain". VISA does 56,000 tps. So it's really simple. How are you going to hit 56,000 tps on the blockchain. The Bitcoin community, which is far bigger and generally more experienced than Ethereum's just spent 18 months vigorously debating this question. Do you suggest Ethereum has come up with a better answer? And the criticism here is around the selling point of Ethereum: "everything goes on the blockchain". You're telling me after 18 months of debate, the best you've got to say about that is: "an account based architecture"? What is it about widgets fully encapsulated in a blockchain-native smart contract on Ethereum that makes it so much more scaleable than a similar widget on Bitcoin's blockchain?
  - 11y ago
semiel 11y ago
Currently, yes. However, the Ethereum foundation is actively working on addressing that limitation, and there are some very clever proposals for how to get around the problem. Unlike Bitcoin, Ethereum development is very active and the community welcomes major changes if they are worthwhile. I agree that Ethereum needs to solve the scaling problem to be successful, but I think Vitalik and the foundation have a good shot at doing it.
- Taek 11y ago
  "Unlike Bitcoin"?? Bitcoin has a multiple-factor greater mindshare. More developers, more researchers, more conferences, and more overall maturity. There are a lot of incorrect things in this thread.
  - semiel 11y ago
    Bitcoin can barely agree to a minimal blocksize increase. The Ethereum proposals for stuff like sharding would be complete non-starters in the Bitcoin world.
mbgaxyz 11y ago
There are fundamental problems with Ethereum which are not addressed in the article. * No guarantee of transaction execution * Turing-complete smart contracts are slow * Scaling is a moonshot * Oracles break the trust model * Public blockchains struggle to stay decentralized More here: https://makebitcoingreatagain.wordpress.com/2016/02/10/5-challenges-facing-smart-contracts/ https://makebitcoingreatagain.wordpress.com/2016/02/10/5-cha...

empath75 11y ago

That's a political problem with the miners and developers. There's no technical reason they couldn't raise the block size.

ufo 11y ago
You shouldn't ignore political problems when it comes to distributed systems... Additionally, raising the block size cap is only a stopgap measure. Eventually you will reach hard bandwidth or disk storage limits and the only way will be to throw some transactions outside the blockchain.
- nickik 11y ago
  Bishares is a blockchain that can do as many and more transaction as Visa. Their does not seem to be a fundamental problem as long as you accept that Worker(miners) are fast servers with good networks.
vonklaus 11y ago
Is this true? I believe there is a tradeoff but want a more in depth explsnation if possible
- maxander 11y ago
  As I recall (and this may not be the whole of it), the "tradeoff" was that a larger block size would risk getting attention from China's great firewall, effectively splitting the network and rendering Chinese users' funds effectively worthless. And it turns out a set of Chinese users have a controlling interest in the network currently... so yes, its a political problem par excellence. It occurs to me as I write that Ethereum may have similar problems down the line, when platform updates turn out to be direly needed, but would disrupt the operation of well-established and lucrative apps.
  - empath75 11y ago
    Also, the miners get more fees when the transaction limit gets reached.

Hermel 11y ago

The Bitcoin network is not having much difficulty with a couple of transactions per second. It is the community that struggles with adjusting to an appropriate throughput.

vertex-four 11y ago
How do I continue using Bitcoin as a transactional currency if I can't perform transactions with it?
- vonklaus 11y ago
  well, you couldn't. Peter Todd is a btc dev and provided a lot of thinking on the tradeoffs. He stated in an interview (fairly confident in this) a 100x improvement in speed and is in favour of not increasing block size. Should check it out, the intetview was on a podcast on the LTB network on itunes. So you will be able to. He has done a lot of thinking around tree chains and scaling and optimiztion. Btc is limited inamount of coins and came first, its value is as a currency. Eth is the first popular blockchaindb. Coins are needed to pay miners and powet dapps. If a currency cant be traded it idnt worth anything, however sidechains seem to provide a way to convert from one to another
- tommorris 11y ago
  I believe the current answer to that is to make vague promises about how the vapourware known as the Lightning Network will solve all the problems.
Crespyl 11y ago
"a couple" transactions per second is not enough for a global network that intends to have millions of participants.

jnpn 11y ago

The Ethereum foundation has been putting a lot of resources into scaling research. Theres some promising proposals involving sharding the validation process, in addition to the off-chain scaling protocols that Bitcoin is pursuing like the lightning network.

pjc50 11y ago

Isn't that inherent to blockchains (every node must have a copy of all transactions)?

jsprogrammer 11y ago
You don't need to keep all the transactions. The original paper addresses how to compact the chain. Basically, you only need to keep hashes.

heliumcraft 11y ago

It's not meant to be used for 'distributed' computation.

eternalban 11y ago
Any system based on distributed consensus is by definition a distributed system.
- heliumcraft 11y ago
  right. But it's meant to be used for consensus and not heavy computations.
  - eternalban 11y ago
    The consensus bit is the "heavy computation". The means regardless of the ends.

davidgerard 11y ago

Every computation runs on every instance of Ethereum, yes. It is important to note that this is not a resource for distributed general computation - it's for running smart contracts. Per the Ethereum development tutorial, "you will not be able to do anything on the EVM that you cannot do on a smartphone from 1999." https://github.com/ethereum/wiki/wiki/Ethereum-Development-Tutorial https://github.com/ethereum/wiki/wiki/Ethereum-Development-T... This is not a very good article - it's rather breathless and short on technical detail. In particular, calling it "the World Computer" is completely not what Ethereum will be capable of.

vonklaus 11y ago
Ethereum refers to itself as this on the website, ethereum.com

crazydoggers 11y ago

One key difference with Bitcoin is that, unlike Bitcoin, Ethereum has an account based architecture. In other words, you have an account that is credited or debited, and you then have a current balance for that account. If you need to get your balance or send ether, you just look up a single number. Bitcoin on the other hand uses an "Unspent Transaction Output" architecture. Basically your balance is made up of all of the credits and debits (inputs and outputs) you've made with that account. So sending bitcoins requires referencing "unspent outputs".. which may be lots and lots of different transactions. That's one difference, and there are others, such as the time taken to confirm blocks (current block time on ethereum is 16s). But there are still lots of scalability issues to be addressed, which are actively being worked on. Vitalik Buterin has a great couple of posts on some of the issues: https://blog.ethereum.org/2014/09/17/scalability-part-1-building-top/ https://blog.ethereum.org/2014/09/17/scalability-part-1-buil... edit: Not sure why this was getting downvoted. Just to clarify, in case it seemed like I wasn't answering the parent. The points above are one of the ways in which "every operation must be verified by every node" is less costly on Ethereum. Obviously if it takes less compute time per transaction, you can scale to larger numbers of transactions per second. Ethereum has already passed Bitcoin in this regard, and has hit at least 8 transactions per seconds at one point, while Bitcoin's ceiling is 7 transactions per second. In addition, prior to launch I believe they managed to push the test network up to 25 transactions per second.

Anonobread 11y ago
But Ethereum's raison d'etre is "X for the blockchain". VISA does 56,000 tps. So it's really simple. How are you going to hit 56,000 tps on the blockchain. The Bitcoin community, which is far bigger and generally more experienced than Ethereum's just spent 18 months vigorously debating this question. Do you suggest Ethereum has come up with a better answer? And the criticism here is around the selling point of Ethereum: "everything goes on the blockchain". You're telling me after 18 months of debate, the best you've got to say about that is: "an account based architecture"? What is it about widgets fully encapsulated in a blockchain-native smart contract on Ethereum that makes it so much more scaleable than a similar widget on Bitcoin's blockchain?
- 11y ago

semiel 11y ago

Currently, yes. However, the Ethereum foundation is actively working on addressing that limitation, and there are some very clever proposals for how to get around the problem. Unlike Bitcoin, Ethereum development is very active and the community welcomes major changes if they are worthwhile. I agree that Ethereum needs to solve the scaling problem to be successful, but I think Vitalik and the foundation have a good shot at doing it.

Taek 11y ago
"Unlike Bitcoin"?? Bitcoin has a multiple-factor greater mindshare. More developers, more researchers, more conferences, and more overall maturity. There are a lot of incorrect things in this thread.
- semiel 11y ago
  Bitcoin can barely agree to a minimal blocksize increase. The Ethereum proposals for stuff like sharding would be complete non-starters in the Bitcoin world.

mbgaxyz 11y ago

There are fundamental problems with Ethereum which are not addressed in the article. * No guarantee of transaction execution * Turing-complete smart contracts are slow * Scaling is a moonshot * Oracles break the trust model * Public blockchains struggle to stay decentralized More here: https://makebitcoingreatagain.wordpress.com/2016/02/10/5-challenges-facing-smart-contracts/ https://makebitcoingreatagain.wordpress.com/2016/02/10/5-cha...