I have a new column today on OnlySky. It’s about a better world that’s coming, and that may not be as far off as you’d guess. The technologies we need to create this world aren’t sci-fi; they all already exist. They just need to be put together in a single package. When they do, civilization is going to be radically transformed.
In the near future, ultra-cheap renewable energy is going to drive out expensive, polluting fossil fuel. Advanced agricultural robotics will take over the jobs that once required grueling human labor. Automated manufacturing will end sweatshops and allow every community to make the things it needs for itself. Electric mass transit and self-driving cars will bring about the end of car culture and suburban sprawl. When you put all these pieces together, what does the completed jigsaw look like?
Read the excerpt below, then click through to see the full piece. This column is free to read, but paid members of OnlySky get some extra perks, like a subscriber-only newsletter and the ability to post comments.
At the top of the list is solar power. Solar (and renewable energy in general) has two crucial traits: it’s incredibly cheap—now far cheaper than fossil fuel—and it’s available everywhere, which makes it inherently decentralized. These facts point to a radically different future than the world we’re used to, where oil, gas and coal have to be transported over long distances from where they’re dug up to where they’re burned.
There are other transformative technologies in the pipeline as well. When you combine them with abundant clean energy, you can glimpse one possible future, like catching sight of a distant valley through a gap in the clouds. Call this possibility the solarpunk future, after the name coined by literary and artistic dreamers.
” self-driving cars will bring about the end of car culture”
You had me going there for a minute. Why not throw in transporters and lightsabers while you’re at it?
i know it’s fun to rip on the musky one, but human drivers cause so much death, i have to think it wouldn’t be wildly difficult to get ai cars to where they can do better.
on the flip, i can’t imagine those who love cars losing interest because options changed.
That’s where I’m at, too, with regard to self-driving cars.
Elon Musk is a lying narcissist; that doesn’t mean the technology itself can’t be made to work. It just means someone more trustworthy should be in charge of it, and it needs to be rolled out more slowly and cautiously, rather than rushed out in half-baked form to pump up stock prices. (Waymo seems to be doing a little bit better.)
In principle, robot cars have all the advantages over human drivers. They’ll never drive drunk, fall asleep at the wheel, get distracted by looking at a phone. They can have reaction speeds that our brains can never hope to match.
Why do you have to think that? Presumably because you (a) have no actual idea of how difficult it is and (b) haven’t bothered to look around you.
Ten years ago, I was more or less where you are. I had this conversation with my wife – I said that by the time our then-yet-to-be-conceived kids were old enough to learn to drive, they wouldn’t have to bother, and by the time THEIR kids were old enough to drive, it would be illegal – there’d come a point where half the cars on the road were self-driving, but ALL the accidents were caused by human drivers. Somewhere at that point governments would step in and tell us we were no longer allowed to get behind the wheel. It made sense. I was pretty confident of that prediction at the time – driver aids in cars were getting more and more useful (I’d just experienced adaptive cruise control and lane-keeping for the first time in a hire car), and many companies were talking up the “real soon now” delivery of full self-driving vehicles. I looked forward to a golden age of motorcycling with the end of SMIDSY (“sorry mate, I didn’t see you” – ask a motorcyclist…), and reading a book or snoozing on the commute or similar.
And then ten years went by.
And dozens (hundreds?) of companies are racing to get full self-driving vehicles to market. Many of the world’s best minds and a great deal of the world’s best money are being thrown at the problem, and regardless of what you “have to think”, the evidence of reality is that getting an ai car to do better than a human is really fucking difficult, even with the vast resources available.
For instance: if you’d asked me in 2015 which I’d see sooner – a dealership down the road from me selling a reasonably-priced family car that could drive itself, or a rocket the size of a 23-storey building LANDING under control back at the pad that it had recently taken off from after separating from its payload in the high atmosphere – I’d have assumed you were joking. The answer would have been so obvious as to be ridiculous. And here we are.
The OP says
… then in the next paragraph talks about self-driving cars, which anyone paying any attention KNOWS to be very much NOT something that “already exists”. It undermines the entire article rather badly, because while the average person may not be deeply into the weeds about insolation rates at high latitudes, battery energy density, conversion efficiencies and other engineering challenges facing this utopian vision, the bloke who empties my bins knows his truck doesn’t drive itself. And a self-driving bin lorry, if you think about it, would be one of the easiest use cases – fixed routes, slow speeds, human supervisors around all the time.
Naive utopianism that opponents of doing something real about climate change can point at and laugh is counterproductive.
Advanced agricultural robotics will take over the jobs that once required grueling human labor. Automated manufacturing will end sweatshops and allow every community to make the things it needs for itself. Electric mass transit and self-driving cars will bring about the end of car culture and suburban sprawl. When you put all these pieces together, what does the completed jigsaw look like?
Among other things, permanent double-digit structural unemployment. And quite possibility, chronic political instability and civil strife among young people with no jobs, no solid connection to the broader money economy, and no legitimate outlet for their frustrations.
We will need to change society in a very foundational way. Current society is based on the idea that people do work that is supposed to be useful for keeping society running, and are paid so they can keep themselves going. At the same time, people derive at least some level of meaning to their lives from the paid work they do for much of their lives. Once AI+robots can do the bulk of this stuff we will need a different model, both for how people get what they need (and desire), and what kind of meaningful activities they’ll be doing.
I wrote a whole book about this scenario. Just sayin’.
Writing a book about it doesn’t make the problem disappear.
I also disagree with how you’re portraying community “resilience.” Resilience to climate change doesn’t only mean shorter supply chains. When the hurricanes come, what happens next?
In the world you describe there is LESS long distance transportation capacity, not more. While you might not need to bring tankers of oil **in**, you will certainly have to bring many people **out**. When the storm surge rips out the first floor of your home and your electric appliances are dangerous and unusable if not completely unresponsive, how do you eat? WHAT do you eat, since your food supply is also local and also affected by the storm.
I don’t deny that there is some form of resilience described in the secondary article (the one linked from within your SolarPunk one excerpted here), but your example is of a ransomware attack. All the infrastructure exists, there is simply none of the usual fuel to run it.
That’s not what climate disaster looks like. California’s wildfires aren’t going to leave homes and automated greenhouses untouched. When an inland wildfire broke out in central British Columbia during a heat wave for the ages, an entire town was destroyed in minutes.
I’m not saying that we will not see benefits from integrating new technologies, but I don’t actually believe your vision is well thought through. As another example you argue that eliminating car culture will eliminate the need for roads, garages and parking lots. And while it will eliminate much of the need for garages and parking lots, the roads themselves will still be necessary for your self-driving pods. It takes so little effort to delete just that one word, “roads,” and thus remove the internal inconsistency from your vision that I can’t help but think that you may have thought *plentifully* about this new society, but you have yet to think *critically* about it.
I’m not pessimistic about any of the technologies you encourage, but you’re going to have to put in a different kind of thinking and then put forward a different kind of argument if you want me to believe your vision has any hope of coming to pass.
Yes, I agree that many settlements currently exist in locations that, because of climate change, are going to become untenable for humans to live. People will have to move away from the coasts as the seas rise; people living in chronically wildfire-prone areas will have to relocate. This falls under the heading of managed retreat, which I didn’t mention because it wasn’t the topic of my article.
On the subject of car culture, it wasn’t an accident that I mentioned roads. Getting rid of (some!) roads is one of the primary benefits of ending car culture.
One of the big problems we have now is, because of the assumption that cars are the only way to get anywhere, every town has to have a road network reaching every house and business. This consumes enormous amounts of space that could otherwise be used for humans. For example, this post points out that the amount of space needed for a large four-way intersection could otherwise accommodate a hundred two-bedroom apartments:
https://www.mrmoneymustache.com/2023/04/07/car-free-cities/
If we rezoned these places for proper density, we could get rid of many of these unnecessary roads and replace them with homes and businesses.
Obviously, in my solarpunk utopia, there will be roads connecting distant settlements, and probably ring roads encircling the perimeters. But the interiors of these settlements would be reserved for foot and bike traffic, with mass-transit options like subways to help people traverse larger distances. In emergencies, paths that are normally reserved for pedestrians can be repurposed to temporarily allow vehicular traffic (usually there are movable bollards to prevent this). This is the model that’s being adopted by cities all over the world that are creating car-free zones.
@3 Raging Bee — Indeed. A society where order can barely be maintained only through fear of extreme repercussion. Something like, I don’t know, fascism?
Or alternatively, a post-scarcity post-work society?
Doesn’t matter how cheap solar gets, it’s not going to supply our winter heating needs up here at 56 degrees north.
Not directly and locally, but… well, interconnectors exist, and of course synthetic fuels (from hydrogen up) can be synthesised from solar elsewhere and shipped the same as conventional fuels.
(Also, total insolation is what counts; sunny days vs cloudy days very much matter, latitude aside)
Newfoundland and Labrador is on the 56th parallel north, and it has a net-zero renewables plan. So is Denmark, and it gets over 80% of its energy from renewables. So is Scotland, and it generates more energy from renewables than it uses – it’s a net energy exporter.
Obviously, solar panels aren’t going to work in places that experience 16 hours of darkness during winter. But solar will work quite well in those same places during summer when there’s 16 hours of sun. And when solar doesn’t work, wind usually does. There’s no place in the world that can honestly claim that fossil fuels are their only option!
Then there is tidal and geothermal power. Even more powerful, in principle.
Still nascent technologies (one more than the other) and not yet price-competitive, but as with other renewables the fuel supply is free, vast, ongoing, and shan’t run out.
(Nukes are far pricier and more problematic)
—
Grid-scale storage is also a nascent technology.
So very many options that don’t rely on chemical batteries, too…
(So, yes, hope yet remains for a better world, because even today energy poverty is a thing)
> Newfoundland and Labrador is on the 56th parallel north, and it has a net-zero renewables plan. So is Denmark, and it gets over 80% of its energy from renewables. So is Scotland, and it generates more energy from renewables than it uses – it’s a net energy exporter.
====
Let’s please stick with the facts. Scotland is not a net energy exporter, not even close. Scotland is a net **electricity** exporter, but it still uses wood, pellets, fuel oil, and similar sources for heating in a great many buildings, and it certainly uses fossil fuels for the vast majority of its boat, ship, train, aircraft, lorry, and car transport (both cargo and passenger).
Can Scotland become a net exporter at some point? Sure, I’d be happy to believe that. I’ve looked at some math for other places — not Scotland to be sure, but other places — and it seems to pan out just fine.
And that’s why it’s so discouraging to read careless statements of yours where you swerve vastly away from the facts to bolster an argument that could still be made just fine if you stuck with those facts.
The original concern was about winter heating, not electricity. While heat pumps can function in colder temps than previously accessible, Scotland uses a wide variety of sources for winter heating, some of them fossil fuels and others (like wood) technically sustainable, but in practice undesirable for various reasons (like the effects of particulate matter on human and animal health).
Further, you’re reducing the relevant dimensions to only latitude. Yes, it’s the only dimension mentioned in the original comment, but your efforts only describe latitude, as if it’s true that if **any** settlement can thrive at 56 degrees than **every** settlement can. But Edmonton is at 53 degrees north latitude and has many days each winter when air-source heat pumps cannot function, while ground-source heat pumps have history of facing extra problems there. This isn’t to say that there are no successful ground source installations, but this is yet another technology that still needs reinvention before it becomes practical. (TBH, thermal energy storage + district heating will probably work much better for Edmonton and other high-latitude, inland communities than heat pumps. The city itself is already working on district heating, though it doesn’t have immediate plans for thermal storage of renewable energy, which makes sense given that it’s still in the pilot-project stage of technological development.)
The point here is that you ran with the easiest version of their assertion to disprove (the idea that heat pumps cannot work at 56 degrees north). But the commenter specified that “we” wouldn’t be using renewables to heat their community, and what “we” looks like changes quite a bit between Orkney and Edmonton. You could easily have steel-manned their argument before addressing it: you did the opposite.
And again, I’m not opposed to your vision, but we won’t get there if we lie to each other about the ease of achieving it (or asserting that Scotland is already there, energy-wise).
Actually I should have checked further.
I don’t know where you got your numbers, but while Scotland might be an exporter of renewably-generated electricity, it’s not even generating all the electricity it needs from renewables. According to the Scottish government itself, Scotland’s electricity is generated by the following sources in the following proportions:
https://electricityproduction.uk/in/scotland/
Renewables: 79.1%
==> Wind: 75.1%
==> Hydro: 3.2%
==> Other: 0.8% (solar is only 0.1%)
Other low-carbon (mainly nuclear): 15.7%
Fossil fuels (mainly gas): 5.2%
So apparently what you’re citing is something like, “Scotland exports 20% of the renewable electricity it generates,” or some such statement. This ability to export renewables rather than curtailing capacity and thus wasting renewable electricity entirely is a good thing, and I’m glad it exists. But that doesn’t make Scotland a net energy exporter.
It seems you need a deeper understanding of exactly what your sources are saying before your next article.
Easy enough to quote the cited article upon which the claim is based:
“In 2020, Scotland had 12 gigawatts (GW) of renewable electricity capacity, which produced about a quarter of total UK renewable generation.[4] In decreasing order of capacity, Scotland’s renewable generation comes from onshore wind, hydropower, offshore wind, solar PV and biomass.[5] Scotland exports much of this electricity.[6][7] On 26 January 2024, the Scottish Government confirmed that Scotland generated the equivalent of 113% of Scotland’s electricity consumption from renewable energy sources, making it the highest percentage figure ever recorded for renewable energy production in Scotland. It was hailed as “a significant milestone in Scotland’s journey to net zero” by the Cabinet Secretary for Wellbeing Economy, Fair Work and Energy, Neil Gray. It becomes the first time that Scotland produced more renewable energy than it actually consumed, and demonstrates the “enormous potential of Scotland’s green economy” as claimed by Gray.[8]”
Scotland (where I live) generates more electricity from renewables than it uses, but that’s less than 25% of our total energy use. Heating is approximately 50% of our energy requirement, which is almost all provided by natural gas, and transport is around 25%, which is almost all petrol and diesel. (https://www.scottishrenewables.com/our-industry/statistics)
Now, certainly, those numbers could be shifted quite a lot of we could retrofit the entire housing stock with decent insulation and state-of-the-art heat pumps, but that’s an enormous undertaking even before you consider the age and condition of that housing stock. Nearly 20% of our houses are pre-1919, which presents significant challenges because of the type of construction, and 25% are tenements, which makes co-ordinanation between multiple owners a big issue. (https://www.gov.scot/publications/scottish-house-condition-survey-2019-key-findings/pages/4/) Heck, I’ve been trying to organise something as trivial as gutter cleaning between 13 tenement flat owners for the last 6 months without success, so the idea of trying to organise a sensible whole-building retrofit doesn’t bear thinking about… It would probably be more cost-effective to demolish and rebuild, but when you’re taking about 20% of the total housing in the entire country, that’s clearly not exactly straightforward.
But even assuming we could do all that (and replace all of our transport with EVs while we’re at it), we would still need to at least double, if not triple, our renewable electricity generation from its current (already quite high) levels to cover all of our energy use. I’m not saying it’s impossible, but we do need to be realistic about the scale of the challenge and the level of investment that would be required.
I do note that in less extreme latitudes, AC for cooling is becoming ever more desirable and even necessary.
I myself live at around 27° S, and I don’t need to heat my dwelling. I need to cool it.
(And it is quite sunny here)
Let me weigh in as somebody who has spent around 200.000€ on that bright solar punk future.
First of all, solar power is nice and a good and necessary element of changing our energy production, but it’s not THE solution. Yeah, multi source punk doesn’t have the same ring to it, I admit.
The problem with all renewable energy right now is that we’re lacking sufficient battery capacity to store energy when it’s abundant for the times when we need it. There are very hopeful projects that could be a breakthrough soon, making batteries that can store enough energy to get a household like ours through the winter. Only that our solar panels don’t produce enough energy to fulfil all our needs.
If we only talk about household appliances, that’s easy. We need around 10 kwh a day, and we produce that amount for more than half the year. We have a small battery block that allows us to be self-sufficient for quite some time even though the sun rises late and settles early. Right now, that battery actually costs energy to maintain.
Not included in all of that is heating and transportation. We have a wood pellet furnace, with all the issues they have. Back when we renovated, heating pumps were only possible with floor heating and that was not feasible. The most important step here is to reduce the primary need by insulating the hell out of this. The US with it’s cardboard houses that cost half a million are a nightmare in that respect, European houses are easier to insulate.
Now, can solar power heat the house: nyes. Our neighbours have a heating pump and solar panels, they “save” their surplus energy in a “cloud”, which is, of course, nonsense. In reality they sell their surplus electricity for nothing in order to have a guaranteed low price when they need to buy it in winter. Again, the actual issue is storage.
None of this accounts for transportation. We also switched to an EV. Driving carefully, the EV need 15 kwh/100km which means less than 2l of gas in primary energy, but when you compare it to everything else, that’s a hell lot of energy.
What we need here is good public transport, bike lanes, and also people having the time to use it. In small cities, a car will always be faster than the bus. People who are already at their personal limit don’t have 1 hour extra every day to use the bus instead of driving, which gets me to the other much discussed point: autonomous driving.
That EV I mentioned? It has an autonomous driving level 2. Highest available in Germany is level 3, which only allows you to take your hands off when in a traffic jam on the autobahn. At level 2, the car has many supposedly smart assistants like an automatic speed limit recognition, stay in lane assistant, collision warning, etc. In reality, those systems are sometimes nice, but often pretty wrong. Where a human eye is easily capable of seeing some street signs, my automatic system is still stuck on the speed limit it read 5 km ago. Multiple lines on the ground in a construction zone? No problem for the human, the assistant doesn’t know which one to follow. And my favourite one is the collision warning: when you drive around a bend an there’s a car parked in the bend, the warning system firmly believes that I’m going to drive straight into the car. First time this happened I almost did because I was so startled. Nope, nope, nope, I don’t trust my car more than myself.
And none of that even touches on the ethical issues. Sometimes there are no good decisions. The famous example being a kid running into the street and you can now either hit the kid or swerve and hit the grandma on the sidewalk. What will you do? What will the AI do? In order for a computer to make that decision, they have to be fed data that tells them which life is more valuable. Do we really want a world where we officially regulate whose life is worth more? Didn’t we have that whole discussion about Covid and disabled people a few years ago? I’d personally rather have a human make a split second decision than a computer coldly deciding to hit person x instead.
In my opinion, the biggest problem with working towards a utopian society of any kind has never been the technology, but the people. There are large swaths of humanity who are affronted at the prospect of becoming on the same level as “those people” (whoever they consider to be their inferiors) that any change in that direction causes them to be knives and guns out, chanting “blood, blood, blood” and looking around menacingly. At which point the “reasonable” and “sensible” people who aren’t directly targeted are all like “Welp, we gotta appease those murder freaks so that they don’t decide to come after us too! But only, like, halfway.” You know, the usual: racism, sexism, classism, ethnicism, religious supremacy, compulsive cisheteronormativity, that sort of thing.
It’s not like that can’t eventually be overcome, given enough corpses on all sides and a long enough timeframe, as we’ve managed to *partially* nerf the effects of *some* of that with all of the various social and civil rights movements since WWII. It’s just that unequal social, political, and economic access strongly tend to result in unequal technological distribution as well. So even if the technology were already here, there is no cultural imperative in most places, little political will, and legions of ones who would rather see the world burn than to allow “those people” to live in a world which guarantees all of them comfort, safety, and health. That could maybe change rather quickly, given factors uncertain or unknown in their effects on the future (or perhaps black swan events), but otherwise we’re looking at a long, slow grind where utopia can be won only with many more decades of blood, sweat, tears, and unnecessary deaths once such a thing becomes truly possible.
One thing which I’ve noticed is that the Political Right in general currently seems hellbent on forcing social policies which they see as coercing people of the nationally predominant race to breed up as many new members as possible (whether or not it would realistically work)… while also being equally hellbent on forcing economic policies which would reduce both the available employment for all and the survival abilities of the unemployed (regardless of whether they understand any likely effects of said policies beyond “more money for the rich in the short term”), and I don’t think they’ve made any real connections between those, because otherwise I can’t imagine what their endgame is.
On the issue of self-driving cars, most technologies follow an S-curve in development. The first stage is “experimental”, definitely not ready for prime time, and generally not available commercially. Where it is commercially available, it’s little more than a “beta test” toy for well-off experimenters and bleeding edge enthusiasts. At this stage, it improves very slowly, taking anywhere from several years to several decades to reach the next step. The next stage is the “takeoff”, where things improve very rapidly in a short time in comparison to the first stage, and is generally available to early adopters. Of course, depending on exactly what the technology is for, early adopters might not be members of the general public, but corporations, government, or military. The final stage is “mature”, when it becomes widely available and commonly used, and development slows down but continues to incrementally improve over the next several years to several decades before finally hitting a wall or being replaced by something new.
Of course, it doesn’t always go that way. Computers, for example, have been very gradually improving ever since the 1939, and have never had any clear “takeoff” phase (lots of mini-takeoffs for certain new niches, but no such phase for general usage). There are occasional forgotten technologies which hit a wall or got replaced before ever getting beyond the experimental phase. (For a fairly recent example, there was a big hype over using microwave weapons as a form of “nonlethal” crowd control during the early 2000s. But by 2010 the whole idea was dropped due to unavoidable physics issues which made it impossible to work as promised.)
Assuming self-driving vehicles follow the usual S-curve development, it’s actually not all that surprising that it hasn’t hit takeoff yet. I remember how in 2008 a lot of very smart people were saying that solar and wind would never be practical for anything other than niche use, and that the energy of the future would be new forms and methods of nuclear power, with Thorium being seen as the most promising; but solar and wind started takeoff a decade later in 2018, while as of the time of writing (late 2024) Thorium reactors still aren’t there yet. Maybe self-driving cars will be in the true early adopter stage by 2026, like a lot of optimists predict. Maybe it will require a lot of breakthroughs, some of which originated in other industries (as it often happens) and it won’t be until 2071. Maybe it takes longer but not that long, but then some black swan tech which pops up in 2039 results in the whole idea of self-driving cars get abandoned much earlier than anyone would expect, perhaps even before truly widespread use. You never know how these things might go, so… just don’t count your chickens before they hatch, I guess.