LLM “personas” seem somewhat useful for tuning output based on need, I.e. senior SWE doing code review on an established codebase vs. vibe coding proof concept demos, but you’re still right… you’ve only narrowed down the set of stochastic outcomes, not collapsed the output to a true senior SWE
Rather disappointing as an environmentalist. But the established freedoms of the car culture and price sensitivity in NA (probably parts of Europe, idk) hard to overcome — “what if I want to road trip though!”
That being said, I think Ford’s shift to a range extended EV makes sense for the truck space. I’m sure someone has crunched the numbers on emissions but getting more market share on hybrid/plugin/range extended EVs are definitely better then ICE only. Plenty of manufacturers are offering hybrids- however, the government has historically been too heavily lobbied to push for hybrids by default and reduce ICE only uptake with some kind of sin tax.
Plug-in hybrids also have tax advantages in some jurisdictions.
E.g. in Georgia (US), EV owners have to pay a $234 annual alternative fuel vehicle fee.
Plug in hybrid owners may choose to have a alternative fuel license plate or standard license plate. If you opt for the standard plate, you don't have to pay the alternative fuel vehicle fee.
Wait, you pay extra for driving a cleaner vehicle when you live in that region? I'm not sure if I understand it correctly or if that's the full picture. Do EVs pay less for road taxes, emission fees, or get some different rebate that this fee is meant to balance?
It’s an attempt to get revenue that would otherwise be collected as gas tax. In practice it’s usually about twice as much as a gas car would generate in tax proceeds. Even in EV friendly states like the PNW.
Ah, I think I understand the difference: we pay taxes for electricity (it's ~35ct/kWh atm) and so there's plenty of proceeds for electric cars as well. Afaik for you it's normal to pay less than half, but then this flat fee for EV owners seems strange when everyone else (in the world, and those driving combustion vehicles) pays per actual usage
We tend to structure taxes with specific targets, rather than rely on general funds for everything. The idea being that public service funding is somewhat targeted towards those parts of society who take the most use of it. Gas taxes traditionally are how a good chunk of our road maintenance is paid for (a good bit comes from the general fund, too, but usually the gas taxes are restricted so they can only be spent on road-related expenses).
EVs throw a wrench into the plan, and so the flat fee is one currently popular attempt to even out the taxes amongst road users. Another idea that got floated was tracking mileage on all cars every year and then levying taxes based on that. But this gets shot down pretty quickly because people perceive it as government tracking of their movements, and that is unpopular.
Personally I think we should just make commercial trucks pay all of it. They already have the infrastructure and policies in place to collect mileage-based taxes, trucks do the vast majority of damage to the roads they regularly travel on, and taxing them would spread around the tax burden to all the citizens who benefit from the existence of the road network (i.e. you get goods shipped on roads, you ought to contribute even if you do not own a car). Local roads should predominantly be funded through property taxes IMO.
Seems like in other parts of the world pigovian taxes are way more popular. They are extremely unpopular in the US. AFAIK gasoline is largely the same price wholesale across the world, but Europeans (as an example) are completely okay with paying more than twice as much at the pump and so more than half the retail price is a pigovian tax.
The difference is that, in many places, the gas taxes are specifically earmarked for road maintenance.
Of course, a better solution would be to pass legislation properly funding road maintenance from the general funds, and raising income taxes to support that. But raising income taxes, even on those for whom a 50% increase in income tax would mean zero change in their actual lifestyle, is politically anathema in these benighted times.
In some cities with HOV lanes on expressways having an EV (with plates) lets you use those lanes with only one person in the car. How important that is to any given buyer obviously depends on where and when they're driving.
It used to entitle you to use the HOV lane in Georgia, but it no longer does.
You also have to pay an additional fee for it as it's a non-standard plate.
I've already been roadtripping across Europe multiple times in an EV that needs 20 minutes to recharge per ~3 hours of driving (e-GMP). To me this is great, and wouldn't be faster in an ICE, since I need the breaks anyway. EV charging is unattended, so it takes less of my time than refuelling. The chargers are already everywhere, even in rural areas and in the freakin' Alps.
I know some people want a pee-in-a-bottle cannonball run, but that doesn't make sense to me. At distances where charging time starts to add up, flying is already much quicker.
Battery swap. Great for industry growth, but the manufacturers selling into US and Europe decided not to go there and disputable claims "it can't work" and false applications of Gresham's law are used to explain why.
The Gresham's law thing: money is just a transfer token. Batteries have a use value. The agents who could profit from hoarding good batteries, don't get to achieve the income of renting them.
It's working fine for scooters, and in China for cars and trucks.
Everyone is now betting on solid state getting both range and rapid charge.
The incentives for such a swapping system are completely busted.
Think of existing swap infrastructure out there, like propane tank swaps. People already use these systems to rinse defective or expired tanks all the time, and that overhead simply gets built into the price.
Now imagine if you could refill a propane tank at home by just plugging it in to your wall. The only reasons to use such a service are now exceptional cases like travel, or to move defective items.
For every new tank introduced to the supply, on average, how many good-for-good swaps will occur before the supplier gets a defective one? Take the cost of a new one and divide it by that average and that is the minimum overhead for a swap.
For batteries, that number is likely in the hundreds of dollars.
An EV with a degraded battery and miles per charge is still very useful for retired people who are tired of traveling long distance and can plug in at home. There should be a good market for them.
I tend to keep my cars over 200,000 miles. Today's cars last a long time. Still, looking back over the past three year's expenses between 150,000 and 200,000 miles, almost all of them relate to engine peripherals - like a new exhaust system, work on emission controls, and a new gas tank, which an EV doesn't have - or brakes, which on an EV last much longer.
Battery swap would just about eliminate planned obsolescence in cars- EVs should last just about forever if the batteries aren’t impossible to replace. The only real wear items in an EV drivetrain are a few cheap bearings that already will probably last a million miles. This would devastate the auto industry.
What about the interior? There’s no one thing you can point to in a car interior and say “that’s no good after X years” but we’ve all seen the interior of old cars.
The batteries in modern EVs will last the lifetime of the vehicle [0] but what factor(s) determine that lifetime are unknown I think.
You can build a car interior that pretty much lasts forever, but it’s not cheap. There’s plenty of 60s and 70s Mercedes around with MBtex and zebra wood interiors that still look pretty much new after 6 decades of regular use. It’s eerie how good those old Mercedes interiors are, I’ve seen cars in the junkyard that were almost completely crumbled to rust powder, but the interior still was flawless. A good upholstery shop can also rebuild a car interior.
Most EVs can have the battery swapped in less time than it takes to drop in a new engine. The real issue IMO is reliable availability of refurb batteries at a reasonable price, and reliability at least on par with an engine.
I’m not sure that is true. I’m a fan of classic European cars and a 70s or 80s Volvo or Mercedes is still a reliable and practical car. It does not seem likely that current battery tech is such that current EVs will have full range on the same batteries in 50 years- in my experience modern lithium batteries often loose 20 percent capacity in just 3-5 years even if you carefully manage them and limit full charge and discharge. And it does not seem like battery replacements are practical or affordable.
I just don’t see current EVs lasting like good quality ICE vehicles do when properly maintaining, because of battery aging.
I'm not sure that tracks with my experience, plenty of 20-30 year old cars still running around. The people upgrading every few years do it for financial or image reasons rather than because the car stopped working.
Warranty anxiety is probably a big factor too, which could be legislated. Imagine how reliable cars would be if a 30 year warranty on drivetrain components was mandatory.
A quality car will last just about forever if you take care of it, Irv Gordon put 3.2 million miles on his 1966 Volvo and it was still in good shape at the end. But most people don’t maintain a car like that, however an EV has the potential to last like that with almost no maintenance.
I don’t think it’s planned (car companies have been competing heavily on lifespan for decades with results) but battery lifetimes seem to already be such that it can last basically forever.
But most people replace their ICE looong before the battery dies. I’d assume the same would happen for EVs too.
That sounds great. We can focus on planning the obsolescence of the auto industry rather than having the industry continue to extract rents on society. Of course this will be a political nightmare, but it does seem like truly the sooner the better.
You are not allowed to decide how everyone else should live. But just for the sake of argument, take a trip to Montana and get back to me on how well universal EVs would work in much of the US. They're fine for urban and suburban areas. They aren't so great for agricultural work, and certainly not great for people who live in places you probably haven't visited. I don't even think EVs are a solution for people in Nebraska, let alone places where the weather gets really extreme.
Our family has EVs and rents a gas car for the occasional road trip. We did the math and it’s cheaper than buying and owning and doing the maintenance on a gas car.
If we did long road trips a lot we’d probably get rid of one EV and get an older gas car for that. It wouldn’t be the daily driver.
Not to mention many companies speedrunning systems of strange and/or perverse incentives with AI adoption.
That being said, Welch’s grape juice hasn’t put Napa valley out of business. Human taste is still the subjective filter that LLMs can only imitate, not replace.
I view LLM assisted coding (on the sliding scale from vibe coding to fancy auto complete) similar to how Ableton and other DAW software have empowered good musicians that might not have made it otherwise due to lack of connections or money, but the music industry hasn’t collapsed completely.
Yep DAW’s aren’t the comparison. People are not thinking deeply about what is going on - there is a big war on-going in order to eradicate taste and make it systematic to immensely benefit the few.
It may just be me, but it seems like digital clocks have been replacing analog clocks long before smartphones.
That being said, I’ve enjoyed wearing my analog watch the last few years to tell time instead of having to pull out my phone all the time. Imagine if we used our Steam decks and Switches the same way…
Almost all wall clocks sold here are analogue ones. And they are in almost every home and public place. (Looking at the clock section in the hardware store and the municipality Office/train station combination)
Most digital clocks are alarms. As soon as it becomes a computer and not primary a time device it becomes more digital.
> Imagine if we used our Steam decks and Switches the same way…
How so? Like, using them as clocks? If a Switch fit in my pocket and showed me the time when I pulled it out without even having to press the power button or interact with it, I guess I can't see why I wouldn't do that?
To be fair, the one thing Musk has going for him is bringing funding to opportunities and making the most of it.
I wouldn't say that Musk is the only person who could have brought SpaceX and Tesla to where there are now, and certainly there are many individuals who contributed heavily to get them there. That being said, not many people have the money and interest to do it.
Bezos had the money and opportunity to do the same as SpaceX but hasn't been even remotely as successful. He technically started Blue Origin first, but wasted several years not taking it seriously as a rocket company; back then it was basically just a space-themed club for him and his friends (Neal Stephenson, etc.) SpaceX went balls to the walls and never let up. The day to day operations are run by Shotwell and she deserves enormous credit, but we shouldn't ignore the role Elon played in recognizing her potential, keeping her happy and letting her do her job (usually) not getting in her way. And a lot of the dreamer stuff, all the Mars colonization and Starship stuff, has Musk's fingerprints all over it. Granted, none of that stuff has actually happened or worked, but it has clearly been good for helping SpaceX recruit highly motivated talent. If you're a young aerospace engineer with something to prove in 2010, which company do you go to? The one that is sending stuff to the space station and talking about putting people on Mars? Or the one that hasn't publicly done anything and doesn't talk about anything either? SpaceX was run extremely well compared to Blue Origin, not just in terms of day to day management but also their big picture strategy.
Nobel prizes in Physics and Chemistry tend to be awarded long after discovery. It's part of the process in evaluating the impact of a specific discovery.
IANAL, but I suspect that the IP situation is similar to current uses, such as catalytic converters.
New tech and specific applications can be covered for commercialization, but the general "idea" of using MOFs for adsorption is broad enough that you'd probably only get into legal hot water if you tried to introduce a direct competitor to someone in the market.
On a more serious note, I’m not convinced that solving NISQ scaling with ML can overcome interconnect scaling. Interconnects are needed to make logical qubits, but introduce increased error rates.
Even with better hardware, the scaling problem remains hard. It’s like getting LLM agents thru complex agentic paths and the hard problem of decreasing error rates.
Even the Aptera, which is designed to be super lightweight, can only regen about 10% of the battery (40 mi vs. 400 mi total capacity) with rooftop solar (https://aptera.us/)
Good reminder with respect to the CAFE standards (rip) that sometimes engineering doesn't trend towards what is "good" with respect to SWaP-C but rather what games the current regulatory environment best.
But the average American drives less than 40 miles per day (33 is the current estimate). The relative charge doesn't matter, because what this suggests is that most Americans would be able to go most days without spending any effort or money at all on charging.
America also loves reserving vast swathes of the road for on-street parking, as well as endless fields of parking lots, not to mention that an enormous number of homeowners would rather park in their driveway and use their garage for storage.
I'm not trying to say solar roofs on cars make sense as a default option, but focusing on "percentage of battery charged" is the wrong metric. Most Americans would get by just fine on a relatively modest amount of charge per day, especially if we got over our range anxiety of insisting on massively oversized batteries for the average EV, which drastically increases weight and decreases efficiency.
Double an ioniq 5's miles per kw/h to 8m/kWh, use the maximum possible insolation (1kw/m^2) and theoretical PV efficiency (70%, reality is 30% btw) and you get... 8kw/h@12hrs a day generated and 3.75kwh needed. about double what this impossibly optimistic calculation gives. Half all the inputs and double the consumption and you get a quarter of the power needed.
It's interesting how much people expect an EV battery to be at 100% charged at all times but have you met the sort of person that tries to keep an ICE vehicle 100% fueled at all times? (It's costly and wasteful and a logistics nightmare, before also getting into questions of how dangerous it can be.)
10% of the battery per day, which would cover all of my driving other than the rare long road trip. I have plenty of weekends where I drive 200 miles, but then the battery would recharge over the week when I drive less than 40 miles a day
