Hacker News Reader

> The engineer in me sees the complexity argument as insurmountable.

The programmer in me dislikes complexity too, but if we apply your argument to cell phones vs land lines, we can see that simplicity doesn't always win.

Presuming you mean infrastructure, land lines have the added complexity of all the property negotiations and intrenched interests that are often corrupt, often dysfunctional monopolies and sometimes both. Putting a device on a pole is relatively simple.

Lot of base stations start to be on buildings and those get same complexity. With downside that the building does not really gain anything as signal is going wrong way from them...

Oh except the building owner gain money. The usage rights for towers are generally leased. I’d note the ability to install one tower for a cell that covers a large region is a much simpler negotiation than rights to connect every building everywhere with wires.

> Presuming you mean infrastructure

Why would you presume that? The guy I replied to made an argument about ICEs being more complex than electric motors.

And fwiw, it looks like that post has been edited a ton. I was replying to a relatively simple (and questionable) statement.

Apples and oranges. If in 1980 we kept our phones with us and expected a cable connection even while driving or in the middle of a field, we would think cell phones were a much less complicated solution.

Because the usage modes are entirely different?

If you could only drive electric-based cars around your own neighborhood, the usage modes would also be entirely different and they never would have taken off either. (As, case in point, did effectively happen in the era of electric golf carts. It wasn't until they got real cars with real car range that things were comparable enough to challenge the status quo.)

point point global bipartite graph connectivity with switching is incredibly more complex than RF cells. The last mile is where the complexity in telephones lie. It’s not in the complexity of the (again) science it’s the number of moving parts. Stringing wires between everything everywhere that can never be broken is much more complex than concentrating the wiring to backhaul lines at towers and devices connecting to the tower and negotiating over a cellular switching protocol. By using connectionless RF for the last mile you’ve dramatically reduced the aggregate machine complexity by orders of magnitude and improved the capabilities (truly untethered) dramatically.

In some weird way rockets are simpler than cars and planes ergo Starlink is simpler than cell networks while providing totally global connectivity.

That’s why I’m looking forward to the rocket to mars I was promised !

Just to serve as a translator for this guy who aren't acronym enthusiasts throwing around high-falutin' words such as "bipartite graph" ICE = "Internal Combustion Engine" RF = "radiofrequency"

Dude said: (1) engines are more complex than battries. So that means battrie cars will do really well on the long term, being simpler and all. The complexity makes engines (also known as "frozen water") more expensive to make, less reliable, and more tedious in general. (2) stringing a bunch of wires is complicated, and you ain't got to mess with that noise with a cell phone because it don't need no wire and is therefore simpler

> The infrastructure discussions are simply engineering and investment challenges, which will yield

I agree with this to a large degree, but it's somewhat beside the point. Those challenges haven't yielded yet, and the state of things in the now are what's important when deciding what sort of car to buy.

Few people are going to buy a car that increases the problems in their day-to-day lives on the hope that those problems will be resolved in a few years.

Given that ICEs are produced in the millions and are quite reliable with very low maintenance requirements, I'd say you are overstating the complexity concerns.

Yes they are complex mechanical things by some measures, but really not more so than say, a mechanical clock. Intermeshed gears, cams, cranks and that sort of thing. The materials and durability demands are different sure, but these problems have been solved.

Many of the other things that are complicated in modern cars are the same in an EV: automatic climate controls with flaps and dampers, the air conditioning system itself, which is even more complex on a battery EV that has to manage battery temperatures, antilock braking systems, automated suspensions that adapt to road conditions, lane-keeping, automatic parking, variable power steering, etc.

I’d note that mechanical clocks have almost entirely been replaced with solid state clocks for effectively all the same reasons.

I’d also note you’re only considering the manufacturers perspective. As a consumer of cars, the complexity of the machine results in a long list of expensive regular maintenance, of which almost none of applies to EVs.

Finally I’d note that saying both have a basis of complexity that’s unchanging doesn’t in any way change the point that ICE then have a ton of additional complexity. Just because at our current scale and research the unit cost of the complexity is about as small as it can help doesn’t mean it has a better future. The complexity of EVs will drop as well, and faster, since it’s still an immature technology and supply chain. I think BYD demonstrates clearly that EVs are already cheaper end to end than ICE, and it’s mostly political and infrastructural work. The political side falls apart in face of economics, and the infrastructure side resolves over time just as oil infrastructure did.

I think the real inflection point will be when gas stations start installing fast EV chargers that captures convenience store traffic and conversion better than the pumps, which are already basically break even or worse margins.

> Many of the other things that are complicated in modern cars are the same in an EV

Yet, when you buy used car - do you look at climate system or make sure engine works well?

I look at the climate system. If it works, it’s a good clue the rest of the car has been taken care of.

Interesting. I maintain my car conscientiously, but I pay almost zero attention to the climate system because given the climate in my part of the US, a working climate system isn't actually important enough to me to spend time or money on maintaining.

If the focus is only the machine and its complexity (and perhaps reliability), then yes. However that's the wrong thing to assess and measure. Consumer concerns are s cost; infrastructure pains; and battery concerns.

Cost is self-evident: Manufacturers need to broadly address affordability and they've been slow to do so.

Infrastructure pains: Chargers are not ubiquitous--charger hogs, who want to eek that last 10%--and not every home/owner can charge (apartment dwellers). Pro-electric owners are doing a disservice, by pretending that this problem doesn't exist or that this problem is solved. This is what Hybrids solve.

Battery concerns: By stats, the % of battery fires and the % of ICE fires are on par. This is where pro-electric owners tend to stop their argument, but it's the details that matter here.

When an ICE model has a tendency to start a fire at home--especially without warning--that's a major recall and considered a defect. When a battery is dangerously damaged? How does one even reliably assess whether battery damage is negible or a risk?

It's still somewhat of the Wild West.

All of these are point in time issues. Cost, I would note, is ICE manufacturers trying to preserve Tesla margins. BYD produces cars that are pretty good at a cost point unacceptable to an American car manufacturer, hence the protectionism and continued high prices that are demonstrably already unnecessary.

I don't agree: It's not something that should be taken for granted.

Battery safety issues may be resolved in a satisfactory manner but that's not a guarantee. If the batteries can be made safer, then great. If it requires civil engineering and planning, then that's a bigger battle.

There's also a matter of sunk cost and market forces: ICEs are fantastically complex machines, but also well-known. Most if not all of the edgecases have been worked out over the decades; and the cost to support the old infrastructure and supply chains are all known quantities.

With EVs, manufacturers will need to build batteries or have redundant suppliers; which is also an unknown quantity and cost. You mention BYD, but given how China is dumping EVs and batteries below cost; the cost isn't exactly a useful metric.

One hopes EVs take off for the sake of climate change, but there are real roadblocks that may stall uptake of EVs.

I live in a cotton mill built in the 1800s. It's a historic property registered with my state, and we can't make many changes to it. Even the concrete pavement is designated as "historic" (everyone finds that ridiculous). Given these parameters, we've only managed to install two EV charging stations for well over a thousand residents. It's wholly inadequate for the 20+ Teslas we have on campus.

The problem for me is that I have ADHD. It's extremely easy for me to forget things like removing my car from the loading zone, so there's no way in hell I'd be compatible with going in and out to constantly move my car around. I'd either never get my car charged, or I'd piss off everyone for leaving it parked in charging for days.

Even if the governing body somehow relented and we could install dozens of charging points, it still wouldn't alleviate the problem for me. There's no way I could organize my life around the ceremony of charging.

EVs just don't work for some people. The math might look right on paper, but the shape of reality has more dimensionality to it.

Your point is fair, but this is a problem to be solved, not to throw up our hands and say it is impossible. EVs work for many people and use cases today; those for which they don't are the body of work to work towards solutions.

I recently worked, for free on my own time, to help a condo association to plan and install infrastructure to support 200 charging stations across two buildings with ~100 units. Total time investment was ~8 hours. A few folks paid to plug into that infra, but those buildings are now future proof.

If you want help, I can reach out; let me know. We will electrify all the things, because we must [1].

[1] https://news.ycombinator.com/item?id=41453394

How about another problem: supply of electricity and then its price? For those of us who don't drive but use electricity, we are going to experience much higher prices of electricity when the masses are forced to drive electric cars. I don't think much thought has gone into planning for this huge demand.

The other problem i see with electric cars is the how long it takes to charge. I'm sure people will respond and say that's just another problem to solve but I just don't see charging ever being as fast as filling up at a gas station. Clearly the activity of using an electric car is going to be different from how an ICE powered car has been used. Has anyone written about this?

> An analysis by KPMG says the U.S. currently has enough generating capability to charge 80 million EVs during overnight hours—hence the need to control when cars are charged. The Edison Electric Institute estimates there could be 26 million EVs on the road by 2030, up from about 3.2 million today. [1] [2]

Most people will charge overnight (either at home or the office), the average daily round trip commute is ~40 miles (1-2 hours of charging at a level 2 charger to replenish this range, depending on current available). Fast DC charging can be performed in 15-20 minutes, depending on state of charge of the battery and the charging station. My Tesla, when supercharging at the grocery store, is finished faster than I finish shopping (for example).

[1] https://assets.kpmg.com/content/dam/kpmg/xx/pdf/2023/07/char...

[2] https://www.wsj.com/articles/url-us-power-grid-electric-vehi...

Thanks for the links. I should have included AI in my Doomsday scenario. AI is very power hungry. Of course AI server farms will not be spread out the way electric charging stations are but will be more concentrated but I do know tech CEO's are already looking at Canada to supply the expected electricity demand. That suggests to me there isn't enough capacity in the USA to meet that forecasted demand.

>My Tesla, when supercharging at the grocery store, is finished faster than I finish shopping (for example). Where I'm at, the number of charging stations at the grocery store is already inadequate. They are always full. I see people with electric cars wanting to "get a spot" with a charger and there isn't any. This is what I mean about the lack of planning. The govt assumes that the market will meet demand but I just think mandating all cars be electric and relying on the market to meet that sudden demand is naïve.

Conversely, it is silly that we simply default to combustion vehicles because they are easy and we are lazy. Work is hard, and nothing in this thread described cannot be solved with existing technology. Maybe we could be less lazy? Maybe!

> hence the need to control when cars are charged

People hate this idea, but this likely one of the biggest improvement we can make on how we utilise excess power.

Imagine everyone plugs in after work, start charging whenever peak ends, either 9PM or 11PM. By 3AM most of the cars are done charging, yet there's 4 hours of off-peak capacity left not utilised. Whats worse, we'll see massive peaks around 9PM if we do not manage charging on wider scale.

Increasing the supply of electricity is trivial, especially today with the advent of cheap and reliable solar. As more EVs join the grid, more capacity will be added, the same as any other technology. Crypto is a great example – Bitcoin scaled up to eating a noticeable chunk of all electricity in the world without anyone batting an eye.

Nobody is being "forced" to drive an electric car – people are buying them because they have much better performance than traditional cars. Quieter, faster, lower maintenance costs, etc.

I spend way less time refueling my car since switching to electric, as many owners will tell you. I just plug my car in about once per week before going to bed.

I think a big barrier is that people used to ICEs want to charge EVs the same way they fill up with gas.

One of the things I had to “unlearn” was wanting to charge to 100% every time. There are many arguments for not doing this, one of which is charging speed.

Another thing I found is that driving an EV makes road trips more enjoyable and less draining because it forces me to take breaks. But it took me a while to fully embrace that.

> Your point is fair, but this is a problem to be solved, not to throw up our hands and say it is impossible.

I don't think anyone is saying it's impossible. They're saying it's a problem right now, and that's a large part of why so many people avoid EVs.

If someone figures out a way to mitigate that problem tomorrow, then the equation changes. But tomorrow isn't here today, and people have needs to met today.

And if the problem is solved and people still choose hybrids or combustion vehicles? You can solve All The Problems and humans can still make irrational, suboptimal choices. What is good enough to make the lizard brain cozy with the EV choice? Not rhetorical! Genuinely interested in the thought.

> And if the problem is solved and people still choose hybrids?

If that's the case, then there's still a problem that hasn't been adequately solved.

But that's a hypothetical. The reality is that there are real, nonimaginary showstopper problems right now.

Here's my lizard brain issue: owning and using an EV requires more planning / is less flexible at least while liquid fuel is so easily accessible.

With liquid fueled vehicles, I can park them wherever at home. If I had an EV, I'd need to pick a spot and always park it there when I want to charge it. If that's inside the garage, I need to keep the garage clean or I can't charge, etc. It's better if I charge with 120v cause I can plug it in anywhere, but then charging is slow. Even if I usually park my liquid fueled vehicle in the garage, I can park it outside for a week or two if I take a big delivery for a group buy and need to have everyone else come and pick it up.

If I move, I have to be picky about where I move; if there's no charging infrastructure, my car turns into a pumpkin.

If I go on a long trip, I need to plan for charging on the way and my destination. From experience with my PHEV, I need to be extra careful because some chargers are impossible to use because their payment system doesn't work; some chargers are impossible to use because they've gone wireless (I seriously saw a charger at a mall that had no cable, wtf).

Driving a liquid fueled vehicle, if the gauge works, I don't need to plan for fueling unless driving through undeveloped areas and there's usually signs like 'no services for 78 miles'. Even without a working gauge, if you keep a gas can, running out of fuel in a developed area is easy to recover from, especially now when you can open a map on your phone and know which way to walk.

There's a car I'm interested in being revealed next month, but it's almost certainly EV only, so it's going to be a no. I don't want to have to think about how to fuel my vehicle.

Also, market rate EV charging out and about seems to be about 2x home electricity prices near me, which means I'd be grumpy if I charged away from home.

>There's a car I'm interested in being revealed next month, but it's almost certainly EV only

Care to share what car that is? Always looking to see what's up and coming.

The new Scout. My first car was a 1978 Scout II, and I'm actually car shopping at the moment, so it could work. This week they said they're going to have a reveal October 24.

I am dumb enough to buy a first model year vehicle. I had the 2017 Chrysler Pacifica (which I ditched earlier this year because of signs of engine failure), and I have a deposit on a Lexus TX (currently under stop sale because of a pending airbag fix), but I'm still shopping if I only have a deposit down.

People still buy mechanical watches despite the fact that a solid state watch tuned to an atomic clock via gps is in every way better. In fact they pay a ton of money for the luxury of a worse device! But as a proportion of all clocks mechanical clocks are effectively non existent.

Watches don’t contribute to climate change.

EV and non-EV cars both do, and the choice to have and own a "classic" non-EV {muscle car | spaceship fin styling | 1930s steam powered racing car, etc.} is similar to the choice to have a "classic" mechanic watch .. which would be the point made in the comment you replied to.

In the overall big picture of timekeeping and transportation the proportion of fully wooden complication watches and Mad Max interceptors would ideally be relatively low.

( but rolling coal will remain popular for some years to come )

And the thread overall is about EVs being drastically simpler machines give them an insurmountable long term advantage to the point that ICE will become anachronistic, and hybrids are even worse than ICE for their complexity. The environmental angle is a non-sequitur.

I would note that EVs are much more likely to have a minimal environmental impact over time as the requisite is electrical storage, not a specific way to store electricity. As electro magnetism is a fundamental property of the universe the ways we achieve that are limitless. Carbon based fuel is pretty much the only meaningful ICE option so there’s not much to improve on there beyond marginal efficiency gains. (I mean I guess we could use liquid fuels like oxygen and hydrogen but that seems unlikely). N.b,. Hydrogen fuel cells are still electric vehicles.

> I would note that EVs are much more likely to have a minimal environmental impact over time

That would be the ideal goal but we (globally) have good way to go yet.

> I mean I guess we could use liquid fuels like oxygen and hydrogen ..

Pragmatically it'd be hydrogen (maybe) in transport trucking on established routes (to minimise infrastructure) and green hydrogen products (ammonia | methanol) for bulk carrier distance transportation to generate central station power (in, say, European winters) for use in EV's.

There's no Engineering (civil | mechanical) upside to hydrogen liquid|gas in mass numbers of small personal vehicles - too slippery to handle with stringent safety requirements to avoid issues with a billion small deployments.

> a solid state watch tuned to an atomic clock via gps is in every way better.

It's not better in every way, or that's the only sort people would buy. It's better at timekeeping, but timekeeping isn't the only reason why people buy watches.

How in the world do you remember to put gas in your car? Are you getting stranded on the side of the road with no gas all the time? Don't drive? How in the world do you manage anything else in your life? You are painting a picture that having to remember (or be reminded) to do anything in your life is 100% impossible.

Putting gas in the car requires taking an immediate action and then being done. Charging an EV requires plugging it in for an extended time and then remembering to retrieve it (when the charger is a shared resource). I could easily believe that the former is easier.

With modern chargers you effectively park at a strip mall, go pee and get a coffee, then get back in your car and continue your journey until the coffee makes you pee again.

Right, which means you have to make time for it. Sure, you can make up stuff to do while you wait, but it's still a long enough amount of time that you have to work it into your schedule. Filling a tank is quick enough that you don't have to do that. You can just fill your tank on an as-needed basis without planning for it.

However, I assure you, the power and telephone company have dispensation to drill holes and run lines. These are regulatory constraints that are basically not real but are convenient functions to induce a behavior, but are easily and summarily dismissed when the political will exists to achieve some outcome. These constrains are point in time and aren’t fundamental like the complexity argument - which is reality based. The constraints you live under may be your emergent reality, but it’s not actually reality.

> cotton mill built in the 1800s

Surprised you are allowed cars then. Perhaps stick to horses and see if state changes their mind.

> The engineer in me sees the complexity argument as insurmountable.

At the same time, the engineer in you/us also needs to understand the circumstances of supply and demand. The completely better technology is unaffordable for a large group of consumers, so the transitional technology is selling better.

From TFA:

"Hybrids are also sort of hurting EVs, aided by competitive pricing that is cannibalizing EV sales. Battery-powered cars and light trucks will account for just nine percent of industry deliveries in the U.S. this year. That’s down from the previous forecast of 12.4 percent.

"some Toyota dealers tell [...] that customers are waiting two or three months just to take delivery of a new hybrid.

"eventually EVs will take over, but customers and infrastructure just aren’t there yet. Until then, hybrids are a fantastic middle ground.

As I said, point in time.

Electric cars have a lot more proprietary systems and complex software that are difficult to investigate. Let alone repair. I personally think ICE vehicles will remain attractive due to proven reliability and repairability.

I sort of agree, but I don't think maxwell has much to do the future of EV's. The silicon and motors have been about as good as they can possibly get for years now. They are only single digit percentages away from being theoretically perfect. If the move to EV's were dependent on them improving, ICE's have a sound future.

The future of EV's depend on one thing: the price of batteries dropping. Once they drop another 50% it's all over. Maybe it's already all over in China.

Batteries are anything but simple. They are effectively nano engineered chemical factories done at massive scale, a scale so big they use 1km sq factories (about 0.4 sq miles) to produce them. Unlike maxwell and silicon, batteries haven't seen decades of cumulative R&D. Right now the press reads like they are at the wild west stage, with new chemistries and production techniques popping up every month.

So yeah, we are at a point in time, but that point in time is being driven by developments in applied chemistry, which is driven by chemists and chemical engineers, not electronics.

Not strictly applies to EVs but Teslas unboxed process is fascinating. Because you need to assemble all car parts separately, you cannot have continuous hydraulic lines hence brakes, steering, fuel, etc are electric making assembly a little bit easier, cheaper and faster.

Legacy auto is falling back to combustion because they are unwilling to eat the short term pain of moving to EVs. BYD faces no such lack of will, and legacy auto will get dragged regardless of this short term bump in hybrid uptake (VW is considering layoffs and plant closures in Germany, which was previously unfathomable [1]). I will make no prediction about Tesla sales, but they are still a contender so long as management doesn't incur irreparable harm to the org; they can also balance between auto and utility scale storage when needed from a manufacturing and revenue perspective. Rivian also has potential, but is not on sturdy ground yet.

TLDR Bet on folks not saving anything for the swim back.

[1] https://www.reuters.com/business/autos-transportation/volksw...

(disclosure: early TSLA investor, no current exposure, know people inside at GM and VW, thoughts and opinions always my own)

"Legacy" auto is falling back to combustion because that's what people overwhelmingly want. The majority of people can afford neither the financial, nor the ability costs that come with electric only vehicles.

Legacy auto can only do this for as long as they don't have to hack and slash to compete against upstarts. US legacy auto is likely to leave China [1], and European automakers will face similar pressures as China comes up to speed as the Clean Tech manufacturer to the world [2]. Global light vehicle TAM is ~90M units/year. Tariffs in the US and Canada are avoided by building in Mexico, for example. China is going to flood the world with cheap EVs [3], broadly speaking. This destroys export TAM for US and European automakers, which will force them to shrink due to revenue destruction.

I would strongly agree that prices for EVs must be driven below combustion vehicles (we cannot expect people who cannot afford EVs to stretch to buy them; price compression and government incentives are material here), but legacy auto must be driven out of the market if they choose to avoid the electrification transition to maintain their profit levels. If I have to argue why (the emergency that is climate change), there is likely very low value in further discussion. Less profits, as many EVs as possible in the hands of people who drive [4].

[1] https://www.reuters.com/business/autos-transportation/detroi... ("Reuters: Detroit Three automakers should exit China, leading analyst says")

[2] https://www.bloomberg.com/news/articles/2024-05-26/xi-s-chin... ("Fast forward to 2024, and China has become the world’s largest auto market and sells more electrified vehicles than any other country, with 9.5 million cars delivered last year. It also controls the majority of the battery supply chain. Homegrown champion BYD Co. dethroned Volkswagen to become the best-selling brand in China and in the last quarter of 2023, surpassed Tesla as the world’s largest producer of EVs. China also overtook Japan as the largest auto exporter, sending 4.14 million units abroad with 1.55 million of them being EVs or plug-in hybrids.")

[3] https://www.bloomberg.com/news/articles/2024-03-02/how-the-s... ("Bloomberg: How the EV Transition Is Reshaping the Global Auto Industry")

[4] https://www.epi.org/blog/uaw-automakers-negotiations/

> Profits at the “Big 3” auto companies—Ford, General Motors, and Stellantis— skyrocketed 92% from 2013 to 2022, totaling $250 billion. Forecasts for 2023 expect more than $32 billion in additional profits. CEO pay at the Big 3 companies has jumped by 40% during the same period and the companies paid out nearly $66 billion in shareholder dividend payments and stock buybacks.

Based on the data, we can afford it, if we stop enriching those folks ^ I just don't care for the lies that we cannot. If legacy auto must die for the EV transition to accelerate or succeed, ¯\_(ツ)_/¯ them's the breaks. These are choices. If legacy auto (both US and EU) makes poor choices, China will force their hand regardless.

Buying or leasing an EV has always been the biggest hurdle for the middle class. With the fed rate cut and soft landing nearly here, I expect we're going to witness an inversion on this hybrid sales trend as affordability finally takes.

The biggest hurdle is being unable to charge a street parked EV; step one is buying a house with a driveway.