3 ms·

I'm more interested in the Daily Transaction Quantity (table on right side http://coinometrics.com/bitcoin/btix http://coinometrics.com/bitcoin/btix ). Bitcoin

by mpg33 13y ago

I'm more interested in the Daily Transaction Quantity (table on right side http://coinometrics.com/bitcoin/btix http://coinometrics.com/bitcoin/btix ). Bitcoin still lags far behind. To put it in perspective: Paypal 7,700,000 transactions per day Bitcoin 67,000 transactions per day.

fragsworth 13y ago
But, by protocol design, bitcoin's number of transactions will not get much higher than that. Only the transaction size will get larger. For a higher number of transactions, you need to include the 3rd parties that provide services as a layer on top of the protocol.
- yen223 13y ago
  If people have to use 3rd-party services, doesn't that defeat the purpose of using Bitcoin?
- oleganza 13y ago
  It's not by design. There was a temporary built-in anti-spam limit (max block size of 1Mb) that is not even hit yet. When the pressure of transactions will push blocks closer to 1 Mb, most miners will raise the limit to not diminish the usefulness of the protocol (and therefore value of their own earnings). Essentially, the limit will depend on network latency. No miner wants to risk creating orphan blocks if his 100 Gb block takes too much time to reach everyone else. I don't expect hardcoded limit to be completely removed. Miners will still want to have some limit to still prevent a situation when someone creates a huge block just for fun, so everyone has to carry it around forever. I wrote in detail on economics of block size limit here: http://blog.oleganza.com/post/43849158813/this-is-how-block-size-limit-will-be-raised http://blog.oleganza.com/post/43849158813/this-is-how-block-... It's a question of whether the hardcoded block size should or should not raised. There's no incentive not to raise it when needed, that's all. Of course, there's also need for some ultra-frequent small transactions. Those will be handled just fine by some clearing houses or distributed clearing networks. All debts will be covered very often with real BTC transaction completely automatically, so there's no much risk of fraud (like creation of yet another fractional reserve system).
  - makomk 13y ago
    Not really. Satoshi may have intended it as a temporary measure, but he underestimated just how hard it would be to change a lot of things about the Bitcoin protocol. In particular, changing the maximum block size will result in a hard fork of the Bitcoin network - since existing Bitcoin clients will reject any block over 1 MB, everyone has to be convinced to move over to the new block size at once. This is unlikely to happen. We've never had a successful hardfork of this kind (I think there was technically a very minor one at one point due to Bitcoin being too dependent on the internals of BDB, but that was to remove an unintended - and previously unnoticed - protocol rule.) Also, at least one of the main Bitcoin developers is strongly opposed to the idea of increasing the block size - he believes that the 1 MB limit is essential in order to avoid a race to the bottom that would end in miners not making enough money from fees, and that small transactions should be done off-blockchain by Bitcoin institutions analagous to banks.
    - oleganza 13y ago
      I don't think he underestimated it. Here's a quote: http://blog.oleganza.com/post/61694565252/satoshi-on-bitcoin-design http://blog.oleganza.com/post/61694565252/satoshi-on-bitcoin... Later in 2010, Satoshi also mentioned that block size limit can be raised if needed in the future. https://bitcointalk.org/index.php?topic=1347.0 https://bitcointalk.org/index.php?topic=1347.0 My point is not philosophical. It's purely economical. When miners start getting much more transactions than they could fit in, usefulness of Bitcoin will become limited by the costs of transactions (bigger fees would be needed to outcompete other transactions and get in the block). If costs go up, value of Bitcoin does not grow or even goes down (because value of money is always speculative: if the future does not look bright, one money quickly loses in value and becomes replaced by some other money). Miners earn bitcoins, not dollars. Investing a lot in expensive hardware, they are very interested in getting a decent return. They would never do something "out of principle" if it hurts their entire business. Block size will be raised by the vast majority of miners from 1 Mb to, say, 8 Mb. It's still small enough to protect against flood, but still gives enough room of growth. Also: "race to the bottom" is just someone's personal fear. In reality, very soon only huge chip factories will be miners. They will produce chips as fast as possible and plug them into their computing clusters right away. Forget about shipping nice boxes overseas, that's too inefficient. Mining will be done by big factories in China or Iceland (cheap electricity). I believe, mining in the hands of small number of big players is not a problem for censorship of transactions or raising their cost. If a miner tries to hurt fungibility of the coins on large scale, he'll simply be boycotted by other miners. (Because they are driven by desire to keep Bitcoin value up.)

by mpg33 13y ago

fragsworth 13y ago

But, by protocol design, bitcoin's number of transactions will not get much higher than that. Only the transaction size will get larger. For a higher number of transactions, you need to include the 3rd parties that provide services as a layer on top of the protocol.

yen223 13y ago
If people have to use 3rd-party services, doesn't that defeat the purpose of using Bitcoin?
oleganza 13y ago
It's not by design. There was a temporary built-in anti-spam limit (max block size of 1Mb) that is not even hit yet. When the pressure of transactions will push blocks closer to 1 Mb, most miners will raise the limit to not diminish the usefulness of the protocol (and therefore value of their own earnings). Essentially, the limit will depend on network latency. No miner wants to risk creating orphan blocks if his 100 Gb block takes too much time to reach everyone else. I don't expect hardcoded limit to be completely removed. Miners will still want to have some limit to still prevent a situation when someone creates a huge block just for fun, so everyone has to carry it around forever. I wrote in detail on economics of block size limit here: http://blog.oleganza.com/post/43849158813/this-is-how-block-size-limit-will-be-raised http://blog.oleganza.com/post/43849158813/this-is-how-block-... It's a question of whether the hardcoded block size should or should not raised. There's no incentive not to raise it when needed, that's all. Of course, there's also need for some ultra-frequent small transactions. Those will be handled just fine by some clearing houses or distributed clearing networks. All debts will be covered very often with real BTC transaction completely automatically, so there's no much risk of fraud (like creation of yet another fractional reserve system).
- makomk 13y ago
  Not really. Satoshi may have intended it as a temporary measure, but he underestimated just how hard it would be to change a lot of things about the Bitcoin protocol. In particular, changing the maximum block size will result in a hard fork of the Bitcoin network - since existing Bitcoin clients will reject any block over 1 MB, everyone has to be convinced to move over to the new block size at once. This is unlikely to happen. We've never had a successful hardfork of this kind (I think there was technically a very minor one at one point due to Bitcoin being too dependent on the internals of BDB, but that was to remove an unintended - and previously unnoticed - protocol rule.) Also, at least one of the main Bitcoin developers is strongly opposed to the idea of increasing the block size - he believes that the 1 MB limit is essential in order to avoid a race to the bottom that would end in miners not making enough money from fees, and that small transactions should be done off-blockchain by Bitcoin institutions analagous to banks.
  - oleganza 13y ago
    I don't think he underestimated it. Here's a quote: http://blog.oleganza.com/post/61694565252/satoshi-on-bitcoin-design http://blog.oleganza.com/post/61694565252/satoshi-on-bitcoin... Later in 2010, Satoshi also mentioned that block size limit can be raised if needed in the future. https://bitcointalk.org/index.php?topic=1347.0 https://bitcointalk.org/index.php?topic=1347.0 My point is not philosophical. It's purely economical. When miners start getting much more transactions than they could fit in, usefulness of Bitcoin will become limited by the costs of transactions (bigger fees would be needed to outcompete other transactions and get in the block). If costs go up, value of Bitcoin does not grow or even goes down (because value of money is always speculative: if the future does not look bright, one money quickly loses in value and becomes replaced by some other money). Miners earn bitcoins, not dollars. Investing a lot in expensive hardware, they are very interested in getting a decent return. They would never do something "out of principle" if it hurts their entire business. Block size will be raised by the vast majority of miners from 1 Mb to, say, 8 Mb. It's still small enough to protect against flood, but still gives enough room of growth. Also: "race to the bottom" is just someone's personal fear. In reality, very soon only huge chip factories will be miners. They will produce chips as fast as possible and plug them into their computing clusters right away. Forget about shipping nice boxes overseas, that's too inefficient. Mining will be done by big factories in China or Iceland (cheap electricity). I believe, mining in the hands of small number of big players is not a problem for censorship of transactions or raising their cost. If a miner tries to hurt fungibility of the coins on large scale, he'll simply be boycotted by other miners. (Because they are driven by desire to keep Bitcoin value up.)