Hacker News Reader

Comment by ggreer | Hacker News Reader

My first thought was, "Is 2 million tons of CO2 a lot?"

According to Wikipedia, "Human activities emit over 30 billion tons of CO2 per year..."[1], so that's 0.007% of yearly CO2 emissions. Germany alone emits 700 million tons of CO2 per year, so that would be 0.3% of their share.[2]

It seems like increasing adoption of electric vehicles would do a lot more to solve this problem. Ground transportation is responsible for 161 million tons per year in Germany (23% of their emissions).

1. https://en.wikipedia.org/wiki/Carbon_dioxide#In_Earth's_atmo...

2. https://ourworldindata.org/co2/country/germany

View on HN

Germany’s problem isn’t the transport sector, it’s the electricity and heat sector.

Compare Germany and France here:

> https://ourworldindata.org/grapher/ghg-emissions-by-sector?t...

While the speed limit would save 2 million tons of CO2 per year, keeping the remaining six nuclear plants running and shutting down coal instead would save 35(!) times more of CO2 per year (70 million tons).

It's also the transport sector.

The cruise ship industry emitted 21930000 tons of CO2 in 2017.

https://www.tourismdashboard.org/explore-the-data/cruise-shi...

And China is worse still. And people cooking non-electrically in Africa, let's get those first.

... or we can tackle multiple areas at once, perhaps? I'm not arguing the Autobahn must be high on the priority list, but simply pointing a finger "this is worse" and then turning away is also not constructive.

Effort is a resource, it takes a lot of work to change things so it absolutely makes sense to have some discipline, create a prioritized list and focus on the things with the most yield.

Too often people jump on the first thing they see and also ignore externalities. How many frustrated autobahn drivers will oppose CO2 reduction efforts elsewhere?

> How many frustrated autobahn drivers will oppose CO2 reduction efforts elsewhere?

Yeah, that is a good point. It must be weighed whether these particular 0.3% are worth it against other potentially contentious issues (and we should additionally consider whether it is likely to become more in the future, or if it's simply a large enough number to matter over the years regardless), hence also my cagey phrasing of "I'm not arguing the Autobahn [in particular] must be high on the priority list".

Still, though, simply pointing at cruise ships and forgetting about potential gains here is not the most constructive thing to do in my opinion.

> How many frustrated autobahn drivers will oppose CO2 reduction efforts elsewhere?

I am not sure how to put this nicely. Autobahn drivers that currently drive at the maximum speed are already quite likely already frustrated. They probably would like to fly a plane and they are constantly hindered by other cars on the road.

In other words, they are the kind of egoists that are unlikely to support any form of CO2 reduction already.

In the end the discussion about the speed limit is less about the environment but more about human safety as a much higher degree of accidents with high speeds end deadly.

Also from personal experience with some of the few highways in Germany without tempo limit ("most" have limits): If there is a truck going with ~80 km/h which is taken over by a bus going ~100 km/h and you take that over with 150km/h but there is nearly an accident because someone thinks it's a good idea to randomly accelerate to around 230 km/h or so on a not so empty highway you start wondering who thought that no tempo limit was a good idea.

Obviously higher speed translates into higher risk.

But given that less than 10% of all accidents resulting in bodily injuries happen on Autobahns, it looks like there are for more important factors to consider than speed alone.

https://www.destatis.de/DE/Themen/Gesellschaft-Umwelt/Verkeh...

Just because other parts of the streets are "less safe" (in given statistics) doesn't mean you can't or should not improve the situations in other places if it's simple, straight forward and with little drawbacks.

(Which is what IMHO it the more important part in the discussion, is a cap of 130 km/h to low? Should it be 160 km/h?)

As a side note the statistics about accidents is not the relevant one in this discussion. The relevant on would be deadly accidents on highway vs. city vs. country side as well as deadly accidents on highway involving speeds > 130km/h and not involving speeds > 130 km/h.

And yes, if we look at deadly accidents of PKWs (~= normal car, i.e. not bus, truck, bike, etc.) then our country side streets are basically death traps.. But then if we consider they in Germany country side streets are often: Less maintained, seamed with trees, areas where wild animals might cross, used by drunk people going home at 3am, used by farming vehicles, used by people and bicycles, etc. it's not that surprising. And honestly there are a lot of country side roads where due to the road being worn down the maximal speed allowed is completely ridiculous.

Edit: Uhm, How did someone one a bicycle die in a traffic accident on a highway? ??

I've seen more than a few instances of bikers thinking the side of the autobahn is a great place to take the fast route to some destination. Probably their GPS navigation or them being rather light on the entire "being alive" thing.

As your example shows, the speed is not the problem, the difference is. If I remember correctly, some research found the accidents to occur where people drive more different speeds rather than where people drive fast.

Overtaking someone without an empty lane of separation with a difference greater than 50km/h is unsafe any day of the week. You have no time to react if that person decides to overtake someone else, if there is something in front of them that they need to swerve around... on German dashcam channels the people that drive 180km/h+ get excessively angry when someone pulls to the left to overtake a truck and I just keep wondering if they have anger issues. I get that they feel like that person almost killed them but they're the ones trying to pass someone with a huge speed differential.

Perhaps we should address the differential if you want to address human safety. Put a speed camera in an unlimited section that just watches the speed difference between adjacent lanes. When too large, a cop can look at the recording and judge the safety of the situation for themselves. That's what they're doing with "smartphone cameras" now in the Netherlands: the camera spots that you are holding a smartphone and takes a picture, then 2 cops independently check the picture to see if the computer got it correct, and then you get a fine in the mail (which you can fight if you disagree).

It seems like increasing adoption of electric vehicles would do a lot more to solve this problem.

Only if you do not get the electricity from burning coal and only if you can manufacture the cars in a way that the reduced emissions over their lifetimes are not offset by additional emissions during manufacturing and decommissioning.

Some more numbers for comparison, 2 Mt are 5.1% of Germany's passenger car motorway traffic emissions, 1.4% of Germany's transportation emissions, 0.27% of Germany's total emissions.

> only if you can manufacture the car in a way that the reduced emissions over its lifetime are not offset by additional emissions during manufacturing and decommissioning.

This is a really bad faith argument I hear, a real eye roller. It never advances the conversation, it is a proverbial “just asking questions.”

Why bad faith? You, your friends and the things you read never bring this up about electric trains and busses. If a Tesla doesn’t amortize by some napkin analysis, neither would an electric bus by that same exact analysis - just because someone somewhere doesn’t like Tesla, doesn’t mean there’s any truth to “electric car emissions don’t amortize.” Just because some article online somewhere says electric busses do and Teslas don’t also doesn’t make it true.

The worst thing the environmental activist community could do is lose Tesla as an ally. Nobody is going to give a fuck about some cobalt mines somewhere, as horrible as they are, or whatever more substantive grievance it is, if we have to endure another 20 years of excess emissions because environmentally conscious people second guess the purchasing of electric cars.

It is a simple fact that using electric cars is no good if you get the electricity from fossil fuels. And I remember reading that at least some electric cars are not reducing greenhouse emissions even if they are charged with non-fossil electricity. If one does not solve those two issues then we are better of continuing to driving with combustion engines. Also I said not a single word about Tesla.

No it's not true. E2e emissions are still lower

> It is a simple fact that using electric cars is no good if you get the electricity from fossil fuels.

Source? Where did you pull that from? An entire power plant dedicated to maximizing power generation efficiency is somehow equally efficient to a car engine that fits in like a cubic meter?

> And I remember reading that at least some electric cars are not reducing greenhouse emissions even if they are charged with non-fossil electricity.

I'd suggest trying to find a link for this too, because people need to be aware of it if it's true.

Just because it might be "in bad faith" doesn't make the argument wrong. A lot of subsidies are being pumped into EVs on the assumption that they're a net benefit to CO2 goals. If they are not, or if it is only true for specific models, then that needs to be accounted for.

> If a Tesla doesn’t amortize by some napkin analysis, neither would an electric bus

That's not a given, these are two quite different vehicles with quite different usage scenarios. A Tesla might not actually get a lot of mileage in its lifetime, whereas a commercial bus would get as much mileage as possible.

Having said that, even if it were true that electric buses also are a net negative on emissions, then the obvious answer would be that we shouldn't subsidize those either.

The argument is also just plain wrong though. Even if 100% of the power generation is from coal, it's still more efficient compared to individual petrol engines in cars. Large stationary constant (ish) speed fossil fuel generators will be more efficient than a bunch of little non constant speed petrol engines in cars.

That's not the argument though. Batteries require a lot of resources to produce, to the point where production of an EV can create more emissions than a comparable gasoline vehicle. If such an EV then fails to offset these emissions during its lifetime, it was a net negative.

Tesla's goal is to build cars that last for a million miles. Electric motor technology is already there and they are also working on making the other parts of the car last a similar length of time. Also notice how the new model S looks very similar to the old model S. Having old cars look like the newer version is also part of getting people to use cars for a million miles/

> Tesla's goal is to build cars that last for a million miles.

The average American drives 13,500 miles per year. Are you sure their goal is to have our grandchildren drive our old Teslas?

If they get automatic driving, you will be able to rent them out via a Uber like service (although I think Tesla's terms of service for their full self driving software package states you can only rent out your car using a system owned by Tesla). I'm sure a large number of people would do this while they were at work or overnight. Especially if it covered most of the cost of owning the car. With this scenario you could get to a million mile in a decade pretty easily.

PS. I'd love to own my grandparent's GTO, Cadallics, etc..

Also after transmission loss, storage loss, losing the 'free' cabin heat and everything? I was under the impression it would at best be a wash.

Regardless, I see your point, and it's easier to switch one big coal plant than try to convince a million people to buy a new car. Better get those cars trickled in and work on the generation side in parallel. Just curious if this fun fact is really true.

I actually think your argument is pretty bad faith.

You are, on nothing but assumption alone, assuming you know this persons intentions, you know their opinions, where they got them, and other people just like them with their opinions.

Then you go on to ignore their actual argument completely, and say, essentially "who gives a fuck if we have another 20 years excess emissions if people end up in cars"

Well, presumably the OP does, because he was making a point as to why it is questionable that EVs are good for the environment.

You basically insulted his character and his argument by ad hominem, then said "who even gives a fuck? I just want everyone to own a Tesla!"

> Only if you do not get the electricity from burning coal

And you mostly don't

> only if you can manufacture the cars in a way that the reduced emissions over their lifetimes are not offset by additional emissions during manufacturing and decommissioning

Luckily you can

> Luckily you can

At present, you can't guarantee that. EVs require more emissions during production, which must be offset during their lifetime. An unfavorable scenario would be having a lot of EVs that go under-used because of a trend towards less commuting. If you're planning to do very little driving, an ICE vehicle may well be better than an EV.

I don't think fossil fuels will be economical for much longer. Per area, solar panels are cheaper than window glass. Costs are now down to $2 per watt. A decade ago it was $7 per watt. Lithium batteries are 1/10th the price they were in 2010. These trends seem to be continuing, and faster than experts project. For example: in 2015 a study looked at past estimates of battery prices versus observed prices.[1] They found that analysts were consistently pessimistic about cost reductions. They tried to correct for this pessimism and made the following prediction:

> ...battery pack costs may in some cases be as low as US$300 per kilowatt-hour today, and could reach US$200 by 2020. This cost development is notably cheaper and faster decreasing than I and many others expected.

The actual cost in 2020 was $137 per kWh.[2]

1. https://www.weforum.org/agenda/2015/04/will-falling-battery-...

2. https://www.statista.com/statistics/883118/global-lithium-io...

The emissions required to produce batteries are what puts EVs at an immediate disadvantage against ICEs the day they roll out of the factory. You then install a bunch of solar panels, which require their own batteries to store the energy that you'll use to charge your EV in the evening.

Before you proceed, please calculate how many miles you will need to drive to offset those emissions, compared to just using an ICE. You might save us all a lot of trouble.

My first thought was "what would the speed limit be?"

130 km/hr (80 mph).

So, in addition to the obvious benefits, this speed limit apparently also reduces CO2 emissions. Honestly, not sure why anyone would argue against this.

It's easy to dismiss every 0.3% improvement as not worth it. I remember reading that the Gates foundation doesn't invest in things that don't have a chance of improving emissions by at least a percent.

But the full picture is to look at both sides. To gain this 0.3% emission reduction, what do we lose?

I love the speed-limit-less autobahn. I'm Dutch but now living in Germany, it's one of my favorite things. It's convenient that you often just go with the flow instead of having to look out if maybe you're doing 5km/h too fast. Definitely nice. But also, I really use it (160km/h+) maybe twice a year for a few minutes. It's fun, but only safe when you have at least 2 clear lanes and also damaging to the environment. Mostly I drive ~100 for the regular (non-pandemic) trips (to friends/family; work is by public transport).

So I'd miss it, but not a whole lot. The vast majority of people's trips would get longer by a marginal amount. Is that not worth it? I'm inclined to say it might be, but I ultimately don't know. I'd love a system where you get an allocation of, say, 30 minutes per year (and it carries over if you don't use it up), but that's not feasible. Or if a speed limit of 130 is posted, then that you can be caught exceeding it by any amount once every five years with a reasonable fine (not losing your license) unless you clearly created a dangerous situation (that's covered under separate laws already anyway). When lower limits are posted, it's always for a reason so there you just shouldn't go faster. Maybe that could allow people to do it if they're in a truly exceptional rush, or just want to, but still reduce emissions by virtually the same amount?

Or we just tax fuel at the price that it costs to take it back out of the air. Then it's simply your choice to drive as economically as you like. But that's also unfair if you live on minimum wage and need to commute by car. Or maybe the first, say, 75 liters a month are tax-deductible, same as with other deductibles? That would keep it affordable for necessary trips but still give the right incentive for those that want to waste fuel.

> I'd love a system where you get an allocation of, say, 30 minutes per year (and it carries over if you don't use it up), but that's not feasible.

We have that limit, it is called mineral oil tax.

That is a bad way to limit things as the yellow vest movement demonstrated.

> Or we just tax fuel at the price that it costs to take it back out of the air

Common we can't just tax fossil energy with a bigger amount of energy. That's not gonna solve anything.

Sorry to be naming a brand rather than a generic technique, but they're the only ones I know of: have you heard of Climeworks? What they do is run machines on renewable energy that take CO2 out of the air. (I have a subscription to support it because I think we'll need the tech to be mature sooner rather than later.)

It definitely would be more efficient to just use that renewable energy directly until we're at 100% renewable energy, but we'll still have chemical reactions. Cement for buildings, creating plastics, space launches, currently airplanes (but it looks like electrofuels might become a thing to solve that), there are a bunch of things for which we'll need this unless we have some breakthrough.

Similarly, if you want to run your combustion engine on the Autobahn, I don't see the problem if you clean up after yourself in some way. Causing more emissions while cleaning up, as you argue, is obviously not cleaning up so that's not what I'm proposing.

Another popular route is to buy these credits, where supposedly someone else will not burn down a particular piece of forest because you bought it and so you can say that you reduced global emissions. While better than nothing, I'm not sure those are as effective as something easily verifiable like Climeworks (easy to see if the machine really poops out solid CO2 as claimed, and easy to see if it runs on renewable energy).

It's not that people "want" to run combustion engines or use fossil fuels. The problem is that over 80% of the world energy output comes from fossil fuels at the moment. Taking the CO2 out then requires more energy than we got out of it in the first place.

So we would need to reduce our fossil fuel usage by something like 90% to have enough energy from renewables to be able to extract the CO2 back (that's just new emission, don't even think about recouping the CO2 from over a century of fossil usage). It's not possible unless we drastically reduce energy consumption to pre-industrialization levels, but with 7+ billion people. It's gonna be back to riding horses.

Barring breakthroughs or divine intervention, there's not gonna be a magical 100% renewable energy replacement at current consumption rates. Fossil fuels are the "free" energy our world as we know it runs on, we don't just have a choice, or can tell people to clean up after themselves via taxes. It's just dancing around the problem, not a solution to anything.

Every little bit counts, but it's really disheartening to see just how little it is in the grand scheme of things.

Another point of comparison: the mass of Earth's atmosphere is over 5 quadrillion tons.[1] 30 billion tons is 0.0006% or 6 parts per million.

These are all really big numbers so it's important to scale them down to something more comprehensible.

1. https://en.wikipedia.org/wiki/Atmosphere_of_Earth#Physical_p...

1°C global warming sounds pretty small too, except its impact isn't.

Yes it is a lot. Removing inefficiency from our system means everything including the stuff you love like driving fast without limit.

While your numbers are probably accurate, it's not helpful to dismiss suggestions like this with alternatives.

We need to do everything we can. Every 0.3% chunk helps.

I'm sure if you used a bus and train instead of a car, you'd reduce way more. But hey, 2M is "really big number" so, let's impress people with that one.

Seriously though, the time they spent discussing this could've probably been spent more productively discussing more environmentally impactful measures. This seems like busy work to be able to tell the voters "Look! We're doing something about the problem!".

And by "doing something" we mean, doing something ineffective, and you bear all of the burden, and get none of the benefit.

If you told a compiler guy about an optimization that would speed up code 0.3% he would be pretty happy.

Sure, but in this case all the elephants are still left standing in the room. And it seems pretty clear that buy-in from the general public is not an infinite resource, so these kinds of changes with minuscule impact but significant public perception of the downsides might not be the way to go. I mean, would you be happy about a 0.3% speed improvement for your compiler at the cost of requiring every source file to be < 4 KB? Especially when the compiler doesn't even inline functions yet?

(Disclaimer: I have nothing against paper straws or reasonable speed limits. But I know that other people do.)

OTOH you can see it as a litmus test (or "foot in the door") for whether people would be willing to make larger sacrifices. Every time there's a discussion of something with a bigger impact, people shut it down because they don't like the personal sacrifice required of them, even when we're talking about their elephant contributions. Yet when you try to start small people still object because the impact is too small on its own and they purportedly want larger changes. I see it unfold right here on HN on almost a weekly basis now. I don't know how to have hope for this planet but it's pretty depressing to watch.

Perhaps that is a signal that the personal sacrifice route is a nonstarter. Maybe focus energies on developing more attractive alternatives?

I don't see any reason why that would be remotely sufficient.

There are far more effective "low hanging fruit" to tackle than lowering the speed limit when it comes to reducing emissions.

- Transporting goods by rail instead of truck or plane. And while we're at it, cargo ships are huge polluters because they can save money when using low grade dirty fuel. Disallowing that practice would have a big effect, but that would require the global community to act (good luck with that).

- Increasing the use of public transport and other ways to lower the need to even drive a car, like (partial) work-from-home schemes, and political incentives driving their adoption.

- Increasing the efficiency of roads, first and foremost by getting smarter traffic flow control that reduces congestion (nothing in transport is more wasteful then a car just idling in a traffic jam). But even the quality of the street and type of pavement you use in roads can have a rather significant effect on fuel efficiency (regardless of whether the fuel is stored in a gas tank or a battery).

Reductions of emissions is one important piece of becoming carbon neutral, but not the only piece. Renewable energy sources and other technologies that will take carbon out of the atmosphere again play another important role in the complex solution that is needed. And e.g. with renewable sources Germany might not look too bad internationally, but we're greatly lacking behind our own goals, and mostly for political reasons.

This isn't that. This is about lowering your server power consumption per annum by 0.3% while lowering the performance of the programs running on that server by anywhere between 0% - 30%. If it is 0% then YAY, if it is a 30% decrease then probably not-so-YAY.

When it comes to cars, the result would be some people would trade longer travel times for lower carbon emissions (while a lot of people do not regularly travel at high speeds on the autobahn anyway if ever). Another question is how indirect factors play out, e.g. will this lead to cars lasting longer because lower speeds will put less strain on engines and other components, or last less long because cars will be operated for longer periods of time per trip. Or e.g. lower speeds would also put less strain on the street requiring less maintenance and repairs.

My point, I guess, is that it isn't as simple as it sounds at first.

That doesn't quite work as an analogy. Reducing speed limits leaves the peak throughput of the road the same: a car arrives every 2 seconds. The only thing it changes is the latency, or the amount of time an individual vehicle needs to travel a given distance.

The individual car (the individual program) will take longer to complete a task.

The humans driving the cars experience the latency. I literally don't care how many other cars drove on the same road in the same hour; I care how long my trip takes.

You should also care about your chance of a serious accident. Germany's motorway accident rate doesn't compare favourably with e.g. the UK or France, although it's a difficult comparison to make.

Oh I do. Having driven in Germany for 6 months (and on periodic business trips) and the US for 30+ years, Germany is a dream from risk of accident, serious accident, and fatality risk compared to the US.

This got me curious. I cannot really find reliable numbers only concerning motorway accidents, at least not this quickly, but when it comes to total road fatalities, it appears Germany is doing worse than the UK but a lot better than France: https://i.imgur.com/ZPHb8cz.png [1]

PS: the US had in 2019 about 110 road fatalities per million [2], the UK had about 26 [1], France 48 [3] and Germany 37 [4].

[1] From page 22 of https://assets.publishing.service.gov.uk/government/uploads/...

[2] http://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/8...

[3] https://www.connexionfrance.com/French-news/Road-deaths-in-m...

[4] https://www.destatis.de/EN/Themes/Society-Environment/Traffi...

The purpose is to reduce the chance of the server room flooding (gradually or catastrophically) by some fraction.

As a side benefit, it also reduces the chance of a server catching fire by about half.

To clarify, I was just trying to address the question of whether the magnitude is large, not the question of how to weigh the various trade-offs.

And yet modern software is amazingly inefficient and sometimes can be speed by 10x if not 100x - developer's time usually is valued more than performance and quality (and security) of the resulting software.

Lowering of a speed limit to reduce emissions is a similar trade-off - it would increase travel time - a scare resource for many people. I'm not saying it should not be done, but the cost is far from zero.

http://proebsting.cs.arizona.edu/law.html

Reply on news.ycombinator.com