6 ms·

Leaving the arguable utility and purpose aside, why is consuming electricity for securing traditional finance ok, but not for securing crypto based financial sy

Leaving the arguable utility and purpose aside, why is consuming electricity for securing traditional finance ok, but not for securing crypto based financial system? Isnt traditional systems consuming arguably needless electricity for ATM security, severs, surveillance, transportation, software and its resources that go into securing the system, etc etc? If thats acceptable i wonder why is the energy spent, in the name of PoW, unacceptable?PoW grabs a lot of eyeballs and also lets these paper analysts to easily quantify the consumption. While quantifying the energy consumption of traditional systems is difficult and hence becomes ok..

jonathanpeterwu 6y ago
Bitcoin energy consumption is a brittle argument, our entire monetary system is backed by much more expensive energy usage. Energy usage for security is a trait of sound money.
mancerayder 6y ago
There's a lot of one-liner talk about externalities in this thread above, but I'd be curious if people would show the stats of energy consumption of the rest of the traditional banking system. I'm not an owner of BTC, but I worked managing datacenter servers in the past, and it's like no one has mentioned a single time that there are arrays upon arrays of servers, 10's of thousands, that run company back-end systems. There are servers to manage the servers to manage the servers (stuff like monitoring, automation management systems, etc.), and it blossoms from there. And these places still use mainframes on top of distributed computing. BTC might be giant waste of resources, but it's curious that the comparisons imply that it's BTC versus no energy usage, instead of BTC versus traditional banks with their giant datacenters all over the country and world. Might be worth at least a mention in the comparison?
- varjag 6y ago
  Traditional banking runs all the world's actual financial transactions (likely billions settlements at any given moment), something that crypto hasn't even started scratching.
  - fastball 6y ago
    Right, but is that not merely because cryptocurrency is a nascent technology? The modern financial system has required an insane amount of capital (energy, resource, human) in order to get where it is. It seems very Luddite to think that the pure-digital system is not more than capable of outpacing (in terms of efficiency and effectiveness) a system that involves massive amounts of human capital, when that (humans > machines) has literally never happened in human history in the long-term.
    - varjag 6y ago
      I'm not sure how say Bitcoin payments are going to outpace in effectiveness what is essentially database queries.
      - fastball 6y ago
        The key improvement with cryptocurrency (and what allows it to be the first truly digital currency / financial system) is the trustless nature. Yes, existing banking is just "database queries", which are of course more efficient than transactions on a blockchain. BUT in order for people to trust those database queries, you need to build a monstrous behemoth of energy/resource/human expenditure to maintain that trust. If I just spin up a PostgreSQL database and tell people "hey, I'll store transactions for you – super energy efficient" obviously nobody will use my system (e: without utilizing supplemental systems for trust). Think about how much human/energy/resource expenditure goes into just making counterfeit-proof paper money – humans designing anti-counterfeiting measures, building money printing machines to make good on those ideas, using plastic/metal threads and such in each and every dollar, humans checking your $100 at the grocery store with a pen / UV light / whatever, people making those pens / UV lights, Secret Service agents investing time and money to find and stop counterfeiters, etc. In cryptocurrency, "counterfeit-proof" comes merely at the expense of more clock cycles. The benefit of which is obvious (to me at least) as the world moves towards 100% renewable energy. Here's a graph to illustrate the point. txns is the cost of an actual transaction. trst is the cost to maintain trust on top of transaction cost. Worth noting that at some point (hopefully), both "txns" rows will be entirely renewably-powered, while the "trst" rows have a good number of processes that are not made more efficient with the onset of renewables. [trad banking] txns: # trst: ############################## [crypto] txns: ########## trst:
- imtringued 6y ago
  Do these companies invest into more data centers because the value of a stock went up?
  - ric2b 6y ago
    Conversely, Bitcoin doesn't need more energy (in any significant quantities) when transaction count goes up.
- deegles 6y ago
  I’m no expert but I would wager that all of the world’s financial infrastructure consumes less electricity than Argentina[0] and that it’s growth is fairly linear and predictable. [0] https://www.bbc.com/news/technology-56012952 https://www.bbc.com/news/technology-56012952
  - ric2b 6y ago
    And you make that assumption based on what?
eladv 6y ago
The article deems neither Bitcoin's energy consumption nor the traditional banking system's as "acceptable" or "unacceptable". (e.g. check out the footnotes.) Bitcoin's energy consumption can be "justified" in many ways (if it even needs justification at all!). But "Bitcoin is a Battery" is not a good justification, since it is demonstrable false.

jonathanpeterwu 6y ago

Bitcoin energy consumption is a brittle argument, our entire monetary system is backed by much more expensive energy usage. Energy usage for security is a trait of sound money.

mancerayder 6y ago

There's a lot of one-liner talk about externalities in this thread above, but I'd be curious if people would show the stats of energy consumption of the rest of the traditional banking system. I'm not an owner of BTC, but I worked managing datacenter servers in the past, and it's like no one has mentioned a single time that there are arrays upon arrays of servers, 10's of thousands, that run company back-end systems. There are servers to manage the servers to manage the servers (stuff like monitoring, automation management systems, etc.), and it blossoms from there. And these places still use mainframes on top of distributed computing. BTC might be giant waste of resources, but it's curious that the comparisons imply that it's BTC versus no energy usage, instead of BTC versus traditional banks with their giant datacenters all over the country and world. Might be worth at least a mention in the comparison?

varjag 6y ago
Traditional banking runs all the world's actual financial transactions (likely billions settlements at any given moment), something that crypto hasn't even started scratching.
- fastball 6y ago
  Right, but is that not merely because cryptocurrency is a nascent technology? The modern financial system has required an insane amount of capital (energy, resource, human) in order to get where it is. It seems very Luddite to think that the pure-digital system is not more than capable of outpacing (in terms of efficiency and effectiveness) a system that involves massive amounts of human capital, when that (humans > machines) has literally never happened in human history in the long-term.
  - varjag 6y ago
    I'm not sure how say Bitcoin payments are going to outpace in effectiveness what is essentially database queries.
    - fastball 6y ago
      The key improvement with cryptocurrency (and what allows it to be the first truly digital currency / financial system) is the trustless nature. Yes, existing banking is just "database queries", which are of course more efficient than transactions on a blockchain. BUT in order for people to trust those database queries, you need to build a monstrous behemoth of energy/resource/human expenditure to maintain that trust. If I just spin up a PostgreSQL database and tell people "hey, I'll store transactions for you – super energy efficient" obviously nobody will use my system (e: without utilizing supplemental systems for trust). Think about how much human/energy/resource expenditure goes into just making counterfeit-proof paper money – humans designing anti-counterfeiting measures, building money printing machines to make good on those ideas, using plastic/metal threads and such in each and every dollar, humans checking your $100 at the grocery store with a pen / UV light / whatever, people making those pens / UV lights, Secret Service agents investing time and money to find and stop counterfeiters, etc. In cryptocurrency, "counterfeit-proof" comes merely at the expense of more clock cycles. The benefit of which is obvious (to me at least) as the world moves towards 100% renewable energy. Here's a graph to illustrate the point. txns is the cost of an actual transaction. trst is the cost to maintain trust on top of transaction cost. Worth noting that at some point (hopefully), both "txns" rows will be entirely renewably-powered, while the "trst" rows have a good number of processes that are not made more efficient with the onset of renewables. [trad banking] txns: # trst: ############################## [crypto] txns: ########## trst:
imtringued 6y ago
Do these companies invest into more data centers because the value of a stock went up?
- ric2b 6y ago
  Conversely, Bitcoin doesn't need more energy (in any significant quantities) when transaction count goes up.
deegles 6y ago
I’m no expert but I would wager that all of the world’s financial infrastructure consumes less electricity than Argentina[0] and that it’s growth is fairly linear and predictable. [0] https://www.bbc.com/news/technology-56012952 https://www.bbc.com/news/technology-56012952
- ric2b 6y ago
  And you make that assumption based on what?

eladv 6y ago

The article deems neither Bitcoin's energy consumption nor the traditional banking system's as "acceptable" or "unacceptable". (e.g. check out the footnotes.) Bitcoin's energy consumption can be "justified" in many ways (if it even needs justification at all!). But "Bitcoin is a Battery" is not a good justification, since it is demonstrable false.