Hacker News Reader

Comment by panarky | Hacker News Reader

"Please respond to the strongest plausible interpretation of what someone says, not a weaker one that's easier to criticize. Assume good faith."

https://news.ycombinator.com/newsguidelines.html

Perhaps it's not idle pedantry to observe that many of those who are angry about crypto energy usage couldn't care less about the energy wasted and pollution generated by other endeavors.

It's like an uncle of mine who is outraged that mosques get tax exemptions, but he doesn't care that churches get tax exemptions.

Clearly tax policy isn't a genuine concern for him. He's angry about something else that he won't say out loud.

View on HN

I wish there was a better name for an "order of magnitude fallacy" (but given Roman numerals I can understand why they probably didn't have a fun Latin name for an order of magnitude problem), because we seem to be seeing them all the time right now.

Industrial production of greenhouse gases is nearly an order of magnitude larger than consumer production, but often inordinately the "guilt" burden is pushed to the consumers: Do you have an EV? Have you changed all your light bulbs to more efficient LEDs? Are you Vegan enough?

Here too: Bitcoin alone has risen to an order (or three) of magnitude more energy consumption than Facebook could ever use/do ever use to track people. (Cumulative, the rest of cryptocurrencies only further dwarf Facebook's comparative energy costs.) We can be angry about two things, we can be angry at both, but why can't we discuss the bigger problem first without getting into the weeds of all the smaller problems?

(And of course, the two order of magnitude problems above are inextricably linked: Bitcoin is on track to make many industrial users of electricity look like chumps and dwarf them by an order of magnitude. Nearly all of the consumer-side gains from veganism, LED lightbulbs, EVs has been offset again, if not entirely dwarfed, by cryptocurrency mining.)

People are bad at reasoning things at large enough scales, have a hard time gut understanding order of magnitude problems, so the fallacies keep creeping up, and keep getting weaponized by bad agents (the consumer "green guilt", the consumer "recycling guilt", so many other "demand-side fallacies" that if consumers just bought "smarter", problems would just go away, when really it's the suppliers that are in control).

https://orionmagazine.org/article/forget-shorter-showers/ << This is the seminal essay on this subject, IMHO.

What is a fair method for determining the amount of acceptable energy for a cryptocurrency or a social network to use?

It seems like you implicitly draw the 'fair' line between FB and BTC.

BTC's market cap is roughly equal to Facebooks. If a POW coin used facebook levels of energy at the same value, would you find that acceptable?

I'm not setting any line here. I'm saying that we could worry about the big problem first without immediately deciding where any "line" needs to be drawn. I'm saying it doesn't matter which side of the "line" Facebook falls on because we could start with the biggest problems first, then circle back to Facebook when it's top of the list (not a whopping log_10 of at least 2 (!) difference from the current worst offenders, such as BTC).

I don't get why anglophones are so obsessed with these "drawing lines" and "moving posts", particularly on recent years. What lines are these exactly? What posts? This is very bad heuristics, actually.

> why can't we discuss the bigger problem first without getting into the weeds of all the smaller problems?

People seem to be intentionally conflating some things. A carbon tax is nice precisely because it avoids getting into weeds and is a general approach that applies equally to everything. Some people are proposing this and others are saying "don't get caught up in the weeds." What? Makes no sense. It just reinforces the poster's point: that some people's real motivation doesn't seem to be environmental, there's a strange focus on crypto specifically that makes it seem like they want it to be reigned in for other reasons.

General solutions are great, but when there's a noticeable log_10 distinction between two examples given, why can't we start at dealing with specific solutions to the biggest problem first and worry about the general approach later? It's a 99%/1% problem: we'd potentially get a huge savings if we dealt with the 99% of cases first and worried about the 1% later.

The raw statistics already tell a story that the difference between "BTC" and "Facebook" is greatly exponential. Wanting to focus first on the (much) bigger exponent isn't necessarily a sign that people's real motivation is "conspiratorial" against crypto.

Because an outright ban is misguided anyway. PoW systems incentivize renewable R&D and investment. If you just straight-up make it illegal, you kill a major driving force of change for the better. If you restrict carbon output instead, you just further challenge the industry to find a way to work inside those bounds, allowing creativity and positive externalities to continue to develop.

There was an article just the other day of a coal plant that was entirely restarted for PoW coin mining. PoW systems are not proven to incentivize renewable energy R&D and investment. There's no proven link there. There's no proof that a PoW coin ban would "kill a major driving force of change for the better". Sure, a restriction on carbon output would be such a driving force, but there is no such restriction today and it is wishful thinking to believe that PoW miners are following anything like one in a market where such externalities continue to be unregulated.

It is orthogonal to the issue at hand that PoW is using far too much energy per transaction/per capita/per GDP/per most metrics you want to point to. Not using the energy in the first place is always going to be greener, no matter how much PoW systems invest in renewables and their R&D! I can't see that as "misguided". An outright ban would get immediate results versus a carbon output tax would incentivize eventual results, maybe. That's not misguided, that's just a different perspective, and a different preference on an ideal time window to address the situation versus wishful thinking and "golly gee, sure hope the market eventually figures it out someday".

How is comparing Bitcoin's energy usage with other endeavors an "order of magnitude fallacy"?

Gold mining uses something like 140 terawatt-hours of energy annually, and produces enormous pollution and environmental destruction in addition to that.

That's greater than Bitcoin's energy usage.

Since gold mining is so much more harmful to the environment, maybe we should outlaw that first before we work on smaller problems.

> Gold mining uses something like 140 terawatt-hours of energy annually, and produces enormous pollution and environmental destruction in addition to that.

The most apropos commentary in this entire thread yet.

> That's greater than Bitcoin's energy usage.

For now. Gold's been around for centuries, Bitcoin not even for two decades. It's got a momentum that Bitcoin may or may not match in the future.

> How is comparing Bitcoin's energy usage with other endeavors an "order of magnitude fallacy"?

The "other endeavors" category above was "digital endeavours" and specifically "Bitcoin versus Facebook", which is an order of magnitude difference.

You've introduced an entirely different category from the above discussion. It's maybe an interesting category [1] to discuss elsewhere, but is off topic from the fallacy we were discussing. Thanks for the non-sequitur, though.

[1] Personally, I'm not so sure it's a useful counter-comparison to Bitcoin: Gold is used in electronics and other industrial needs, beyond its service as a value store/commodity of interest to collectors. Gold is also increasingly rarely the "primary" focus of mines. Most of our Gold today comes from Copper mines and very few would argue we are mining Copper as a value store/solely for greed.

> Why do HN comment threads always end up as Pedantry Pageants where all ye who dare comment must address all the fucking edge cases

Gold mining isn't a fucking edge case if it uses more energy and produces far more environmental damage.

According to the principle you described, Bitcoin is the edge case we should ignore until we solve the gold mining problem.

Gold mining isn't even an edge case in that specific example when the class of problem is explicitly defined as "What digital services use too much energy?" as it had been in the above conversation. Is Gold a digital service? No. Is Gold Mining wasteful? Perhaps. (Though again, as an aside, the better question with specific respect to Gold is: Is Copper Mining wasteful? Most gold is mined as a by product of copper mining.)

It's not an edge case of the "electric waste of Digital Services", because it is an entirely different problem. So not only do we have "pedantry pageants" of edge cases, but we get to address the "pedantry pageants" of all the unrelated but seemingly related problems too?

not to mention human exploitation and harsh working conditions, particularly on illegal gold mines

Fair enough. I love this orange website, but I also despise it. Flying off the handle is rarely a sensible solution.

To your point, though, I'm willing to wager that most of the people whose biggest problem with crypto is energy consumption would happily also support other harsh measures against all manner of pollution generating entities. I know I would.

A pollution tax is really the only viable solution though. Playing whack a mole against individual entities and activities will never solve the problem, and creates a lot of legislation.

So long as pollution and carbon are free or cheap, people will find wasteful ways to generate it.

Incentives work, and taxing carbon and pollution will incentive green alternatives.

a pollution tax will not work. That just becomes the price of doing business and that cost is factored into other areas of the business model. Even worse, it just gives more capital to inefficient governments

> That just becomes the price of doing business and that cost is factored into other areas of the business model.

Yah that's the point and business with less environmental impact will be more competitive. Industries/products with more of an impact will be more expensive, reflecting their true cost.

> Even worse, it just gives more capital to inefficient governments

Most pollution/carbon taxes are proposed to be revenue neutral, i.e. all the proceeds are returned to the tax base. This is also important because like other (non-luxury) consumption taxes, a carbon tax is regressive.

I have a few disagreements with the MMT crowd, but on this one point they are transparently correct: the exact level of taxation has little relationship to government ability to spend. Talk of "giving more capital" to governments shows a deep misunderstanding about the nature of fiscal constraints.

> other areas of the business model

And what are those? District heating?

It becomes the cost of doing dirty business.

Solar is taking off not because we banned coal, but because it's cheaper than coal. If coal and natural gas were taxed according to their carbon output, solar would be even more ahead, nuclear may become more attractive as well.

Aluminum is heavily recycled not because it's green, but because it's significantly cheaper than mining new.

Incentives work. The free market responds to incentives, and works around regulations.

In the specific case of proof-of-work cryptocurrency, I don’t think a carbon tax is directly useful: If it’s unilateral then mining trivially shifts across borders; if it’s global then it’s an inflationary pressure on the coin price which offsets the reduction in mining you might otherwise expect.

> Perhaps it's not idle pedantry to observe that many of those who are angry about crypto energy usage couldn't care less about the energy wasted and pollution generated by other endeavors.

Have you observed this, or are you simply speculating?

Was thinking the same thing... There's no evidence I can see to suggest the parent is not also against other forms of excessive energy consumption.

There's a hidden subjective value judgment built into the word "excessive" that's doing a lot of the heavy lifting here. These "regulation" comments tend to have a high likelihood of boiling down "I want the government to stop things I don't like", where environmental concerns (crypto) or data collection (adtech) is a mere pretense grounded in incomplete information at best, emotion at worst.

Personally speaking I find it just as irritating of a pattern here that knee-jerk calls for "regulation" (i.e. bring in the coercive power of the state) or outright banning are so often floated as a/the solution to every problem. When legal coercion is suggested, it is entirely proper to bring up the existence of the dragons that lay down that road.

I want the government to look seriously at things that are using the energy footprint of a mid-sized country simply to spin up a new financial instrument. Particularly when there appear to be equivalent things that don't.

If cryptocurrencies didn't have the energy footprint, I would find them pretty uninteresting and/or ridiculous for various reasons, but ultimately it wouldn't bother me that people do what they do with them. When they start adding seriously to the global energy load in a time of climate crisis that makes them pretty offensive.

it's not simply a "new financial instrument", it's a whole ecossystem

I don't know whether you're referring to me or some of the other replies, but I am definitely against most forms of excessive energy consumption if they can be avoided.

Are there any estimates of how much power crypto mining has consumed and how much running all of the Facebook/Instagram/whatever servers has consumed?

Best i found for Bitcoin was 40-100 TWh per year: https://digiconomist.net/bitcoin-energy-consumption

And for Facebook it appeared that they used 5140 GWh in 2019: https://www.statista.com/statistics/580087/energy-use-of-fac...

So essentially:

  Entity    Consumption (GWh)
  Facebook   5 140
  Bitcoin   40 000 - 100 000

I wouldn't have expected Facebook to consume comparatively so little energy, if i'm doing conversions right.

The argument stands for itself. If your uncle gave a good argument against tax exemptions, it's a good argument for tax exemptions. Just pretend somebody else gave it that doesn't have the same alterior motive.

Attacking the motive or whatever is not said out loud is responding to a weaker interpretation

> Perhaps it's not idle pedantry to observe that many of those who are angry about crypto energy usage couldn't care less about the energy wasted and pollution generated by other endeavors.

I don't think that's correct. I think that it's possible to care about both, but to be dismissive of misdirecting subject changes, i.e. "whataboutism".

So you answered his complaint about the use of "slippery slope fallacy" by deploying the "whatabout" fallacy and created a strawman.

Nicely done, nicely done.

Reply on news.ycombinator.com