6 ms·
This is not a bug, it's a feature.
by gdhbcc 7y ago
This is not a bug, it's a feature.
- realusername 7y agoIt is a feature yes but that makes this technology very inefficient to process transactions.
- sk0g 7y ago"Very inefficient" is an understatement too! I wonder how much energy Visa and Mastercard combined use to process their transactions. Mind you, they probably do orders of magnitude more transactions per second. I recently had a discussion with a friend about the viability of a distributed/ blockchain alternative to YouTube. Extrapolating BTC inefficiency to a distributed CDN processing TBs and PBs of traffic...
- masklinn 7y ago> "Very inefficient" is an understatement too! I wonder how much energy Visa and Mastercard combined use to process their transactions. Mind you, they probably do orders of magnitude more transactions per second. There's about 6 orders of magnitude difference in transaction cost. Mastercard/Visa would use ~2W.h per transaction, Bitcoin about 1MW.h.
- Mengkudulangsat 7y agoBitcoins energy consumption has no correlation with its transaction throughput. Price goes up > People turn on more miners > Difficulty goes up > Less efficient miners become unprofitable > People turn off more miners > Difficulty goes down. It's a self-correcting dynamic purely due to economics, not linked to usage.
- sk0g 7y agoSo ever since Bitcoin's creation, it has been using the exact same amount of power? Because the transaction throughput surely has changed. Do you not agree that, loosely, there is a correlation between transaction throughput/ popularity/ real world usage, and the amount of people being happy to invest time, effort, and money into mining?
- Mengkudulangsat 7y agoThe amount of power may not be the same, but the economic formula hasn't changed from the start. Inputs to that formula are solely bitcoin price / hardware price / electricity price / other miners. Miners don't care about network throughput at all. Is this market failure? I don't know... Bitcoin still works despite it.
- viraptor 7y agoThroughput increased (from 0), since people started using it. But the maximum throughput cannot increase without protocol changes. There are workarounds like lightning network, but that's off the main blockchain. There's a hard limit on how many transactions you can put in one block. More mining power doesn't change that maximum.
- realusername 7y agoI would not go as far as to say there's no correlation. The price of Bitcoin is also dependent on how many people would use it / value it, so the profitability of miners. Multiply by 100 the usage compared to now and for sure the price would go up by a lot.
- paulmd 7y agoWithout this "feature" it would be trivial to double-spend transactions. The entire concept of Nakamoto consensus (upon which Bitcoin security rides) relies upon provable waste of electricity to a degree that the average transaction is impossible (or at least financially unviable) to reverse within some widely accepted timespan. And yes, this implicitly makes Bitcoin a bad idea, as well as blockchain in general.
- zaarn 7y agoWell, any blockchains that uses this amount of energy for it's consensus, though there are plenty of blockchains that don't really need that much energy (Git doesn't use much to run a commit either)
- ulrikrasmussen 7y agoBut git also doesn't solve a distributed consensus problem. The problem of reaching consensus on the official HEAD commit is externalized to traditional PKI methods, e.g. GitHub.
- zaarn 7y agoWell, no, the consensus of a git repo depends on who use it. For example, on Linux the true HEAD is where Linus says it is. Everyone else is wrong. That's a consensus of everyone working on Linux. In more distributed projects, HEAD is where everyone agrees HEAD should be. Human consensus is a valid form of blockchain consensus.