8 ms·
This is completely wrong. They scaled energy with number of transactions. > At each time step, CO2e emissions for the given number of transactions were estimat
by spiorf 6y ago
This is completely wrong. They scaled energy with number of transactions.
> At each time step, CO2e emissions for the given number of transactions were estimated based on the emissions generated to mine that number of transactions in 2017.
Bitcoin network does not work that way. Energy is spent securing the blockchain, not per transaction.
- imtringued 6y agoThe amount of energy wasted depends on how high the block reward and the transaction fees are. The block reward is going down over time and transaction fees are unlikely to grow beyond a practical maximum.
- tromp 6y agoTransaction fees will have to go up to compensate, else Bitcoin will become insecure as its resistance to 51% attacks falls below critical levels. See discussion at [1]. [1] https://bitcointalk.org/index.php?topic=5306354.0 https://bitcointalk.org/index.php?topic=5306354.0
- Rochus 6y agoYou mean peer reviewers of Nature did a bad job?
- matzab 6y agoThis was a Comment (dated October 2018), not a paper.
- Rochus 6y agoSo Nature publishes comments without reviewing them?
- spiorf 6y agoThat assumption is wrong. Take it as a single data point about the reliability of their peer review process. EDIT: I really mean that is a single data point. A usually reputable publisher can make mistake sometimes.
- jevgeni 6y agoYes, see, here we have respect for science only if a) it is our specific field, or b) fits average tech bro sensibilities. Obviously, we are completely objective and not emotional or biased, because we know JavaScript.
- vntok 6y agoIt very much seems like they do, doesn't it?
- gus_massa 6y agoAlso, it was not published in "Nature", it was published in "Nature Climate Change". Now "Nature" (the publisher) has many journals, and many of them has a name that starts with "Nature", but they are different journals, with different teams and different qualities. https://www.nature.com/siteindex#journals-N https://www.nature.com/siteindex#journals-N
- mjfl 6y agopeer review is not infallible. Stop worshipping it.
- Rochus 6y agoDo you have a better concept?
- adamnemecek 6y agoThat’s a terrible response.
- jevgeni 6y agoSo “no”.
- Rochus 6y agoWhy?
- adamnemecek 6y agoYou can criticize something without having a replacement.
- Rochus 6y agoAnd how is that supposed to help us in this case? It doesn't have to be a "replacement", but at least some suggestions for improvement; just rejecting without any suggestions is not constructive.
- benibela 6y agoA wiki Then anyone can correct mistakes
- Rochus 6y agoI wouldn't publish the results of my multi year scientific effort on a platform "anyone" could make changes. Reviewer are expected to have a suitable qualification and experience and no bias. The publisher knows the reviewers and can select suitable ones. In a wiki, where anyone can publish or change anything without control, the scientific level - at least in certain areas - will inevitably level down. There is no perfect system, but there are many worse than we already have.
- hatsunearu 6y agoNot only is peer review not some omniscient god, Nature isn't exactly flawless either. Plenty of top tier journals fuck up all the time.
- Spivak 6y agoYeah “per transaction” cost is calculated by taking total energy expenditure rate divided by the transaction rate. It’s a mistake to assume that the ratio is the thing that’s fixed. Energy expenditure is related to fees and the blockchain reward.
- spiorf 6y agoIt's a mistake to assume energy source composition too. I would guess that a majority of mining is renewable right now (mining is a perfect sink of intermittent excess energy, and because its profitability is related to energy cost, renewables win). More so in the future.
- eloff 6y agoThe capital expenditure is sufficient that it still makes sense to run the miners 24/7 and not try to take advantage of cheap electricity - although that does matter somewhat when deciding where to place the data center. This is true of pretty much every industrial process. The idea of running production only when the sun is shining to absorb excess solar seems like a universally poor idea.
- spiorf 6y agoRarely industrial process can switch on and off randomly and still scale revenue linearly. The capital expenditure argument makes sense when mining is the primary business. Using old and otherwise unprofitable (but cheap) hardware makes sense where there is intermittent excess energy.
- imtringued 6y agoHaving cheap energy 50% of the time is still better than 0% of the time.
- eloff 6y agoOf course, that goes without saying.
- Denvercoder9 6y agoFurthermore, they assume that Bitcoin will follow "the adoption trends of other broadly used technologies", which is unlikely to happen. Even the cryptocommunity is starting to accept that Bitcoin has effectively failed as a payment network.
- xwdv 6y ago> Even the cryptocommunity is starting to accept that Bitcoin has effectively failed as a payment network. Where are you seeing evidence of this?
- SulfurHexaFluri 6y agoI don't have proof for the larger community but I was very interested in bitcoin at 2012-2015 and it really did seem like it was going to be a real alternative, all the big companies were starting to accept bitcoin but now its all gone. Almost nowhere accepts bitcoin anymore because they failed to solve the issue that transactions cost on the level of dollars to process so its only useful as a speculative investment and not a practical currency.
- oivey 6y agoIn hindsight it makes a lot of sense. The cost of a transaction for Bitcoin is always going to be higher than just updating a value in Visa’s database. Visa may have some increased costs due to consumer protections, but consumers are pretty clearly willing to shoulder the cost.
- clcuc 6y agoWeren't Ethereum and Cardano supposed to solve some of those problems?
- SulfurHexaFluri 6y agoI have never heard of Cardano and Ethereum seemed to market itself as a way to run the internet and government inside of blockchain while the reality was it was never used for anything but speculative investment. Cryptos have basically burned the idea and the general public will probably not care of any future solution since all of the previous ones were scams or not very useful. These cryptos all promise the world and deliver nothing.
- rtkwe 6y agoThe number of transactions are limited by block size though so each completed block has a maximum number of transactions it can include.. also securing the blockchain isn't an end unto itself it's only really useful when it includes transactions.
- sp332 6y agoSegregated Witness (SegWit) was activated in July 2017. https://en.wikipedia.org/wiki/SegWit https://en.wikipedia.org/wiki/SegWit And overlays like the Lightning network have been in use for even less time.
- SulfurHexaFluri 6y agoThe true statement is "We consumed this amount of power and were able to process this amount of transactions" but the power usage does not scale with transactions. You could increase the block size and get far more transactions without changing the power usage because the power is not used to process transactions.
- mdoms 6y ago> You could increase the block size It's not that simple. As I recall the last time a block size change was proposed it fractured the entire community and produced a splinter "currency".
- spiorf 6y agoThe splinter fractured and has his own little splinter too now!
- SulfurHexaFluri 6y agoIts technically very simple but politically very difficult. The point I am making is that the power usage is not directly tied to transactions. Not making a transaction on the blockchain will not save X watts because that power is burned regardless of what transactions are going through. An example would be that me not driving to the beach saves some fuel but bitcoin is just a train constantly driving to the beach and back regardless of how much I use it. You could grab the number of trips and the power usage and calculate a power usage per trip but its not an entirely useful metric since the power usage does not scale up as more people use the service.
- Geee 6y agoAnother major error they made, is that they assume that Bitcoin mining uses grid energy and assume that it follows the CO2 emissions of electricity generation in that country. This is completely wrong. Most mining happens currently on underutilized hydropower stations, because grid energy or fossil fuels wouldn't be as profitable. Because of the mining difficulty adjustment and diminishing returns in improving ASIC hardware, the only way to profit is to find cheaper energy sources than all other miners. In the long term, Bitcoin mining accelerates the development of abundant and cheap energy, which is not fossil fuels.
- viraptor 6y ago> In the long term, Bitcoin mining accelerates the development of abundant and cheap energy, which is not fossil fuels. Yes, but does it reduce the total reliance on fossil fuels? If we start with X coal plants and build a Y hydro plants to support btc mining, we're still running the original X coal plants.
- Geee 6y agoBitcoin mining can possibly finance the building of new solar / wind farms without government subsidies. In time, mining will become less profitable and it makes sense to sell the energy to grid. Obviously, fossil fuel energy should be taxed more heavily to get rid of it faster.
- viraptor 6y agoThat sounds like trickle down economics idea. It's possible of course... But is it more likely than people taking out their btc gains and buying a yacht instead?
- dschuessler 6y agoI suspect by "develop" the parent comment did not refer to the construction of power plants but to the improvement of the technology. Like in "Research & Development".
- microdrum 6y agoWelcome to progressive politics. Science and logic do not matter. They will use climate change to take your chosen store of value. Because your chosen store of value reduces their power over you. Pretty simple, really.
- agedcayenne 6y agoThe amount of energy spent does not depend on number of transactions nor how much is needed to secure the blockchain, it's purely determined by how much miners are willing to spend, which is mostly determined by the bitcoin price. So, if you would expect the amount of energy spent on mining to scale with any variable, it would be trading price and price alone.
- jasonlingx 6y agoThis is it. Amount of energy spent = profits from mining / cost of energy. While price is the main factor affecting profits, fees/volume and the constantly decreasing block reward also play a part.
- bryanrasmussen 6y agook, but I can't help but think there must be some relatively negligible energy cost per transaction that nonetheless might be not so negligible when total number of transactions added together - what that energy cost is I wouldn't be able to say.
- wmf 6y agoIt really is negligible compared to mining, like a factor of a million or billion.
- omgJustTest 6y agoCall them out on GitHub, would like to see this refuted if true.
- Animats 6y agoIt's worse than that. The assumption is that Bitcoin transactions eventually become the world's payment system. Then this is multiplied by Bitcoin's very inefficient work per transaction.