Embedded-systems programming is not taught, and no one is willing to pay for training. The result is that development is outsourced to entities that claim, falsely, to have the knowledge. Eventually the consequences of the fact that they do not have the knowledge surface in an undeniable manner, and the only way to cover is to make a great show of a fresh start. (This affects all industries, not just automotive, but right now that is where the spotlight shines.)
I'm not sure this is exactly the problem. It sounds like turning the car into a platform with changeable parts has caused both organizational and technical problems.
To be fair, im still not sold that this is an advancement except maybe in simplifying the number of components. I'd prefer the car to work without "updates" and DLC. Why does my car need a firewall??
It needs two! One to keep engine fires out of the passenger compartment[1] and one to keep unauthorized users or code out of your infotainment and control systems.
It's not practical to produce a car that never needs updates. That would be a bug-free system, which is impossible. Since they're going to ship updates anyway, a lot of focus is on minimizing the cost and hence OTA.
For what it's worth, I work in this industry and the general rule of thumb is that every increase in validation from QM (standard quality) up to the various levels of safety critical code has up to 10x the cost per line of code of the previous level.
But those cars are no longer competitive. There is only a marginal buyer group who wants to drive these "bricks", which would also unlikely pass the requirements set for new cars.
It was not. Recalls have included software updates (sometimes via component replacement) since ECUs became common in the 1980s. Reverse engineering the binaries and flashing updated parameters is actually how ECU tuning used to be done.
Updating the software in the computers that control the car has traditionally been combined with providing diagnostic support for it through the dealerships, not done OTA. Having an OBDII connector has been mandated in vehicles for a long time, you plug something into it that lets you either listen to CAN bus traffic or reprogram an individual Electronic Control Unit (ECU).
Now that all vehicles have entertainment systems connected to the internet, I guess it is tempting to use that to reprogram ECUs, I haven't been working in this area recently though.
The first use case of connecting entertainment systems to a vehicle bus that I can remember was to read some engine settings and turn up the volume on the radio at higher speeds.
> The first use case of connecting entertainment systems to a vehicle bus that I can remember was to read some engine settings and turn up the volume on the radio at higher speeds.
Is anyone actually begging for this though? And why do you need a full bus? This feels like a luxury car problem that could be solved over I2C or something.
I’m reading this whole SDV thing, and outside of using less ECUs, it seems like an overengineered solution to what was hardly a problem. If we can update ECUs already with OBD-II, step 1 is just making a virtualized OBD-II port that the infotainment system can talk to. Everything else can then stay unchanged until later.
One problem is that the ECUs are fairly dumb, they each have a limit on how fast you can send CAN frames to them without overflowing receive buffers. The protocol to reprogram them starts by asking the target ECU how much of a delay is needed between each frame then needs to keep to quite tight timing constraints when sending the new flash image, I have written a Linux network protocol module to do this.
I absolutely enjoy speed compensated volume. It's nice to have about the same apparent volume inside the cabin as road noise increases while not being very loud when going slow speeds or stopped.
I2C is also a bus, just one that's less reliable and involves more custom work to use.
A "virtualized OBD-II" is really just a UDS server if I understand what you're trying to convey. UDS is a dumpster fire of a protocol that should be expunged from existence, but my personal feelings aside can be run anywhere you want. That exists. I'm not aware of many systems that directly connect the infotainment processors directly to critical CAN buses. Usually there's an intermediary component to isolate them.
>That would be a bug-free system, which is impossible.
Yes, but code that doesn't get written does not have bugs. And I don't want to control the rear window defroster, wipers, climate control, fog lights or whatever, on a touch screen menu buried 7 levels deep while going 130 km/h. It's bad enough that coffee makers, light bulbs and tooth brushes now have updatable firmware.
> That would be a bug-free system, which is impossible.
Why? If the rest of the car can function within design specifications for years, why can't the firmware?
I'm fine with updates to add compatibility with new protocols and such, but to me a bug implies there's a standing problem with the current system that's not due to some sort of wear/changing standard/component damage etc. While one can point to examples of cars with defective mechanical designs, I don't think anyone considers it impossible to create designs without such defects (where defects are defined wrt. specifications), why is this the view in software engineering?
The rest of the car doesn't function within specifications for years. That's what recalls are fixing. These days, a lot of software recalls are being issued to work around physical design "bugs". The Tesla cybertruck frunk pinching issues are a well-publicized example.
But, do you have an example of a software project anywhere that's bug-free? I'd include the space shuttle code, but even that famously high quality development process produced a (low) number of bugs.
Not open source but mostly reverse engineered, and automotive: the PCM code for GM’s LT1 engines. The only thing that could be considered a bug is the behaviour of wide-open throttle fuelling, which was completely acceptable from the factory but made aftermarket tuning a bit tricky. Specifically, the fuel calculation routine would use the most recent BLM (short- and mid-term air-fuel ratio correction factory as measured by the O2 sensors) when calculating fuel delivery when wide-open, rather than locking it to a constant; the “most recent BLM” may be for a completely different area of the tune (like mid-RPM low throttle), where things like vacuum leaks or even just intake runner inefficiencies have a much greater affect on AFRs than when wide-open. This can result in either too much or too little fuel being injected, and under- or over-shooting the target AFR.
The reason for this is a physical limitation: the cars weren’t shipped with wideband O2 sensors, so there’s no way to measure the AFR when wide-open (since it’s targeting a significantly richer mixture, and narrowband O2 sensors can only signal whether a the combustion is stoichiometric, or rich or lean relative to stoichiometric, with no further info). The implantation is probably not a bug but rather a compromise; in an ideal world, the “most recent” BLM will hopefully be from an “almost wide open” part of the map, and the general rich/lean characteristics will be close enough. And, the fuel table in the factory tune is quite safely rich when wide-open, so even with a leaking injector causing the idle BLMs to be way off, the fuel being pulled when wide open will still be completely safe.
> It's not practical to produce a car that never needs updates. That would be a bug-free system, which is impossible. Since they're going to ship updates anyway, a lot of focus is on minimizing the cost and hence OTA.
What was wrong with ECU and ABS etc software prior to the OTA era that we're now apparently entering?
I've had plenty of cars--too many--and outside of a few warranty repairs involving re-flashing ECU/ABS(maybe), this was a very rare occurrence.
(Not counting deliberate tunes or re-flashes for modification purposes)
Purely from a manufacturer's perspective (not personal opinion):
One, it's expensive. If your update takes half an hour to apply, under the old model someone's being paid half an hour to apply it. Either the manufacturer cuts the billable hours to the dealer and the dealer loses, or the manufacturer is paying that half hour out of increased prices to the consumer. With an OTA system there's usually no cost to anyone besides network traffic. This amounts to billions of dollars in savings for manufacturers.
Second, owners hate 1) paying for updates and 2) getting notifications about it in the mail. It generates bad press and bad experiences for the manufacturer.
Three, it makes the production line more efficient.
Four, the old systems sucked to maintain and for techs to use. They were also insecure and retrofitting security is impossible in a standards compliant way. The internet people have done a much better job with their standards.
Five, most owners are not like you and I. It's a feature for them that their car gets improvements and fixes automatically.
Six, you can be pretty certain what the rollout distribution is. Regulators don't like it when owners are driving around with years old recalls active because they forgot to schedule a dealer appointment. Manufacturers don't like keeping the inventory around.
Seven, "networked services" can piggyback on the same infrastructure and provide additional revenue streams. Certain corporate types think of this as one of the main benefits. Remember how manufacturers used to sell annual maps updates that no one bought? Some consumers also enjoy these sorts of networked services, which frankly I find a bit baffling.
They are not complex, so less breaking point to begin with.
Also some bugs are considered not critical enough to do recall, they can be fixed when the owner return their car for maintainence. But now even those small bugs will be fixed by OTA update
> It's not practical to produce a car that never needs updates. That would be a bug-free system, which is impossible.
Hmm, I disagree. Bug-free systems are expensive and hard, and get more expensive and harder as complexity increases, but you can absolutely produce a car that never needs updates. The vast majority of computer-controlled cars from the 80s to the early 2010s never needed updates, and the ones that did were performed at dealers (and were usually for non-critical things, because the critical things were simple).
GM had a good run in from the mid-90s to the mid-00s producing bug-free cars, even with some complexity. I don’t know of any software issues on any cars with LT1 or 3800 engines, nor with any of the tech in the Northstar Cadillacs. Displacement-on-demand could be considered a buggy implementation, but it was working as designed, and never got patched out, so I don’t think it counts.
That’s of course ignoring the decades of cars that had no computers at all. No software bugs being patched out with OTA updates in a carburetter (you have other problems obviously though, namely terrible fuel economy and emissions, and generally lower reliability).
If you make it a hard requirement for a car to be bug-free (maybe outlaw OTA updates and force physical recalls on any software problem?) I can guarantee manufacturers can make a bug-free car. It’ll just be way less complex and have way fewer flashy features, and will either cost more or have lower margins. It’s been done in the past, it can be done again.
There is a sweet spot for the level of computerization in cars. We had it somewhere around the year 2000, then waaaaay overshot, and haven’t corrected back.
This is all the more frustrating as I'm in the security side if IT, and have been trying to teach myself C and assembly for embedded development and understanding how malware and vulnerability exist in this ecosystem and how I can help address these issues.
You can find router firmware sourcecode online and find pretty egregious vulnerabilities if you're really trying to learn.
Alot of embedded stuff is outsourced and doesn't want to waste the computing power for stuff like stack canaries. I recall the following from making a tool for a dlink? router?
//Reads a file name
foo ReadFilePath()
{
// Get file name
// TICKET 21321: Fixed crash by increasing buffer size
char FilePath[100];
ReadFileName(&FilePath);
}
It sticks out to me, since the crash was clearly from a buffer overflow, and they had this documented in the source code that increasing the buffer size fixes it. What they didn't realize was that the bug would still happen and you could get a buffer overflow from this and do whatever you wanted. This is the level of programmer you're dealing with who's writing embedded software in an overseas sweatshop. And the talent isn't even there domestically since they're severely underpaid compared to someone writing simple javascript.
The people who actually can do it are not underpaid. These days they are brought in to do cleanup. They can name their price and pick their assignments.
Only if you’re looking for top dollar when you graduate. Which unfortunately a lot of people are. It makes sense, most people pick this field for the salary and not out of passion (like the vast majority of professions).
But if you take a couple C/assembly/systems electives, look for internships at hardware companies, build a couple of toy projects on the side, and graduate with even a modicum of embedded experience, there will be companies that will hire you, pretty much guaranteed. You won’t be making 250k out of the gate, but you should still be making a more-than-livable salary (and frequently in a lower cost of living area than, say, the Bay), and if you pick companies correctly, you can be working with and learning from some truly genius engineers.
The pipeline’s there, it’s just not attractive (read: $$$$$$$) enough to pull in most people in the industry.
I took an embedded course in university where we programmed the AVR AtMega 328p on the Arduino UNO not using the Arduino libraries and compiler. Make files and setting up an environment.
But yea, a single class probably isn't sufficient and also I image a lot of embedded companies have a preference to hire someone already familiar with the chip they are targeting and the toolchain for the stack. I also see a lot of asking for experience with RTOS, which in my class, we didn't use an RTOS.
RTOS-based development varies significantly from RTOS to RTOS, so I’m not sure how much it’d help to learn to use one. On the other hand, most fundamental OS knowledge is fully transferable to RTOS, so that would be helpful for embedded developers to understand.
Yes, there’s a ton of specificity. Could probably say that about kernel dev too. But there is a ton of things people do that’s a lot more generalized. Of course I’ve used very little of specific things I got tested on in my day to day over the years.
Programming embedded devices is not the same thing as "embedded-systems programming". The latter means, first and foremost, that the software is not allowed to crash, ever, for any reason, else it is people's lives.
I did some initial requirements work on a system to monitor continuous-web papermaking machinery; the line had to be stopped, physically and completely, within 100ms if anything went wrong, because an uncontained web of paper can literally cut people in half. They wanted, in order to be able to hire, to use one of the embedded flavors of a well-known consumer-grade OS, and I had to prove to them that there was no way to make any of them safe, at any cost. And they knew their hardware, because they had built it themselves.
The absolute last resort is a watchdog timer that hits the reset button if N milliseconds go by without the software telling it it's okay. This is what you have to implement if you are dealing with buggy and undocumented hardware -- as, all too often, you are. Sometimes you can get some doco for $ and an NDA, but then in order to get the real doco it is much more $$$ and a much tighter NDA, and the existence of that option is not even divulged until after things have already gone very far south.
If it were only a matter of reading the top-level doco for this or that chip, there would be no issue.
Why do the hardware companies make things so difficult?
If I were selling hardware I’d want it to be as open and well documented as possible. So that more people buy it and so that I get credit for all the great stuff people make with my products.
There are a few reasons, using the hardware manufacturers’ logic.
1) The more you open up your design and its behaviour, the more your competitors can learn about your product and how to possibly improve their own. Even stuff as basic as what specific features/capabilities a specific SKU at a specific price point has can be useful information.
2) The behaviour may be sufficiently undefined as to make documenting it impractical (or a bad look). Specs may also be padded (“up to 14 bits of SNR” may mean you’re getting 8 most of the time unless you’ve got a golden sample, and you’re not getting the distribution without paying big bucks and signing a big NDA). This ties in with 1) - if your competitors know your exact yields, they might be able to advertise being better/more reliable more truthfully, or even cheap out on their manufacturing a bit to drop their own yields down to match or just slightly beat yours.
3) The behaviour might be unknown. There’s obviously a crazy amount of validation testing that goes into high-end chips, but even the best test plan can miss things. This is especially true when you’re talking about high-speed stuff and anything involving power delivery/voltage fluctuations, or async/pipeline executions, or a million other things that can go wrong. Again ties into 1) - if your competitor knows that your chip might deadlock the radio with an obscure pattern of inputs and control signals, that could give them insight into how you’ve laid out your silicon and might give them optimization ideas.
4) If all the available info is given out freely, then potential customers can easily compare manufacturers and pick the best one. The manufacturers don’t want this, unless they’re the best, for obvious reasons. And, because everything’s locked down so tightly, no one knows if they’re the best until the chips are on the market and the volume contracts are already signed. And those contracts are hard to break, since the specs agreed upon are pretty vague due to 1-3.
5) The manufacturer knows their chips suck, but needs them moved anyways. This is rarely the case from most non-discount manufacturers, but it can happen. In this case, you don’t want to give away anything you don’t have to, because most info you give out is going to drive customers away to a better option. Good example in the consumer space is Intel refusing to publish acceptable voltage specs for their 12-14th gen Core chips, which resulted in motherboard manufacturers overvolting and killing high-end CPUs to try to meet the frequency specs Intel was advertising. If Intel was truthful in their voltage and frequency specs, there’d be a minuscule percentage of chips that could actually hit the advertised frequency at safe voltages, and 99% would have worse performance than expected, which would almost definitely result in lower sales.
6) The behaviour may be highly dependent on external factors. Basic example, a chip with external DRAM might have its execution pipeline stalled more or less frequently based of DRAM spec, or a wobbly voltage regulator might be known to cause lockups when certain executions are happening. Are you going to tell your customer those problems, or just say “we recommend high-speed DRAM and high-quality VRMs?” Especially if the other guy just says “we recommend high-speed DRAM and high-quality VRMs?”
The world would likely be a better place without such logic, but the incentive is there. Until someone comes and breaks the paradigm, I don’t see things changing.
Yes! But it's also obvious that the industry doesn't have a prayer to ever reform. Stuck between proprietary and NDA's chips everywhere, using proprietary and NDA's toolchains and development kits, to product proprietary DRM'ed products.
This is an industry that is about as far from the light of science & enlightenment as it is possible to get, ensnared as deeply in the entangling anti-human anti-science Intellectual Property qualgmire-hell as can be got. Oh sure plenty of science goes it! It's fantastically interesting & technical! But aside from some Application Notes write-ups trying desperately to help move the practice along, move it out of jank, knowledge goes in, but it doesn't ever come out! There's such a lack of peershios with which to practice science, to report your findings to, to replicate works on.
The software world talks about its patterns and practices. The biggest industries on the planet are building software like wild AND are mad into open source. But... computer engineering is the shadowland, where no talk nor victories that happen there are allowed to be shared, where nothing escapes confinement. What a fucking plagued awful land of people unable to ever do the right thing, unable to bring their work out of the dark & into real civilization.
I can find 1000s of posts or blogs or whatever on every React nuance, Rust thing, LLM trend, or whatever, but nothing even describing what "real" embedded programming looks like in any fashion (I'm not counting blinking an Arduino LED here).
What does writing ABS module software look like? I'd actually love to know--it's not an area where you can "vibe code" your way to a 'working' product.
Can't speak for cars specifically, but other software built to comparable standards requires a super rigorous process including detailed specs, risk analysis and mitigation, detailed design, comprehensive automated and manual testing against the soec including test coverage, tons of documentation on everything and finally a third party certification on everything.
The process is so far removed from typical web and business slop that it's an entirely different world of its own.
Automotive has the problem of overwrought frameworks and no-code tooling that make it hard to fix problems and make improvements. Once the original devs are burned out or laid off the codebase rots and gets handed off to maintenance devs who barely know how anything works.
I'm waiting for a recall fix for the underpowered Sync 2.5 system to correct a backup camera problem. I'm not looking forward to worsening of all the current bugs with USB audio file playback that cause the UI to hang or not show a fully rendered display.
Companies are not willing to pay what the people who know embedded deserve. $150,000, $200,000 and up for a JavaScript webshit "engineer", $100,000 max if you work in embedded, unless you have a super specialist knowledge maintaining software on NASA's remaining PDP-11s or whatever that they can't afford to lose.
Fortunately that is incorrect. I mentioned in another comment, but I’m well over $100k USD equivalent in salary alone as an embedded engineer, working in a relatively low cost of living area in Canada, graduated 3 years ago. Working for a “regular” company.
Maybe things just really suck for embedded in the states? But since my last year of university I’ve been inundated with recruiters for embedded positions, and I’ve never had a problem finding work. ~75th percentile in salary alone for software engineers in my area, ~55th-60th for Canada. I make more than every JS developer I know who graduated with me, except for the ones who moved to Seattle, Vancouver, or the Bay.
My CS degree concentration is embedded systems. I love embedded programming, but it would probably cost me $200k a year to do it versus the backend distributed systems stuff I do now.
Admittedly I don’t know your salary or market, but it is possible to make decent money in embedded. Connections & market timing are both vital though, in my experience, as well as being actually good at your job. I’m in Canada so numbers are way different, but salary-wise I’m in the ~75th percentile software engineers in my area, my title is embedded engineer, and I’m fairly junior (3 years out of university, ~6 years full-time experience). I’m working with some other embedded people who are in the 95th percentile for software engineers in the country. The main problem is there are very few high-paying embedded jobs; conversely though, there seem to be even fewer highly-skilled embedded engineers looking for work. I recently interviewed at a company paying 50th-85th (based on experience) percentile trying to hire pretty much any competent embedded engineer, and their problem isn’t insufficient salary, it’s just a lack of applicants or any skill level. From what I’ve heard, the same seems to be true pretty much everywhere.
Now sure, if you’re looking for 500k+ jobs, embedded isn’t the area to be in, unfortunately. But I prefer low-stress, fun-environment embedded jobs, and don’t mind trading off salary for that. Different strokes.
Problems like this always come down to salary. I love embedded (hardware in general, really) and would absolutely love to do it, but during my entire career, the salaries for embedded have been so much lower than you get for slinging JS/web shit. And now with 15 years in, the gap is even worse.
At this point, when I wanted to get back into hardware, it made more financial sense to outfit my home office with all the measuring instruments, debuggers, tools and other equipment necessary for embedded work and do it as a hobby. If I had the space, I could even get full-size CNC machines and still come out ahead cash wise. It’s insane.
It’s no wonder they can’t find experienced embedded devs, when it makes no financial sense to stick with it over a decade.
it is possible to get an embedded-adjacent job that pays well over 500k (my friend was bit banging FPGAs), it'll just be at an HFT instead of a car company
It’s definitely possible. There are probably a few embedded engineers making over $1M. But those are all outliers. It’s a whole lot easier to make $500k in web dev.
it's a lot cheaper to pay one exec a couple million than to staff a medium-sized software engineering org: even 500 people at an average fully burdened cost of $250k is $125m/y.
I wasn't disagreeing with you; I idly did the mental math, was surprised at how high it was, then thought through how I'd probably finger-in-the-air that it would take a couple thousand eng years to build a whole car OS. You're building safety critical software, so you're going to start with a very serious test effort, etc. So finger in the air a good chunk of a billion dollars a year for many years.
One major issue has been that paying a developer market rates is practically unthinkable to traditional automakers. If you were to apply to a mid/senior job in Michigan, you might get offered $125k. The typical workaround has been to establish "software offices" on the west coast with separate pay scales and separate corporate structures that largely function as internal "external" vendors. The C suite are able to pretend they're not overpaying, and the teams getting work done are able to attract people closer to market rate.
Well, it’s not just about pay scales. The developers just don’t want to live in Detroit (or even Ann Arbor) for the most part. And coastal East isn’t really that much cheaper for the most part.
I’d probably add that the pay scale for software vs. electrical/mechanical people probably wasn’t notably different in the 90s or so. And California rates didn’t compensate for CoL in general. Very different.
Old car is massive amounts of mechanotechnical engineering, with some software for keeping the beast under control and provide some basic entertainement.
New car is basically a computer on a simple chassis with an equally simple drive train. Software and battery tech is everything.
I'd argue that chassis tech is more sophisticated in the BEV case due to more weight. Adaptive dampers, air springs, rear-axle steering, etc. might not be necessary on a comparably sized ICE vehicle.
OTOH, ABS and ESP systems can achieve similar or even better results with less complexity because motor torque control is inherently low-latency, which can also complement brake deployment (hydraulics is not as well behaved as e-motor).
You do get rid of emissions control and tiny little sensors / flap actuators sprinkled all around the engine bay, so yeah, probably overall still a simplification win, but I doubt you can get very far without "massive amounts of [Mechatronics] engineering".
One of these things is not like the others. Tesla, for good or ill, needed to write a full stack for their EV. Not only did they need to do it, but they did in fact do it and ship it and develop it over several years. Recruiting a Tesla software guy is probably the best choice between these three. And he'll cost you less.
Both Google and Apple have car software, and who knows if Apple actually developed a full stack of the way Tesla did. But anyone can download and play with android automotive. It's unclear what getting one of the android automotive developers would do for you.
Whoever convinced the people writing requirements documents for car user interfaces that they needed to use Unreal Engine to show you what your own car looks like and spin it around in. 3-D deserves some kind of salesmanship Nobel prize. That is the most pervasive useless thing I've seen in a long time.
> One of these things is not like the others. Tesla, for good or ill, needed to write a full stack for their EV.
And so did traditional manufacturers, they just had the benefit of being able to phase it in if they so chose. Or they could have done a hard cutover, either way, the failure is on them for ignoring the benefits of the Software Defined Vehicle discussed in the article.
> It's unclear what getting one of the android automotive developers would do for you.
Do they do vehicle control systems or just infotainment?
> they needed to use Unreal Engine to show you what your own car looks like and spin it around in. 3-D deserves some kind of salesmanship Nobel prize.
I mean that's exactly the kind of thing that makes Tesla fanboys rave endlessly about their car. It just needs to be decoupled from the actual software system, like any UI.
Android automotive doesn't come with software for battery management and functions like climate control, headlights, error notifications, and other driving functions. But it does provide the best toolchain, widget set and user interface framework for those functions. It also comes with support for multiple screens, multimedia, multiple languages, speech recognition, app stores, cameras, wifi hotspotting, OTA updates, modes for vehicles in motion, Bluetooth, pointing devices so you don't have to be all touch all the time, etc.
All that stuff adds up. As Volkswagen found out.
Green Hills supports running android in a VM so you can do all of the safety critical things like traction control, and ABS in a secure environment.
I have driven several different, rather new, cars over the last two years. The most hassle-free experience was the second cheapest of the bunch, a 2024 Opel Corsa GS (a Stellantis brand). I actually was sad when I had to give it back.
Now I read that Stellantis is behind on the software game and I wonder if there is a relation. Seriously, I'm all for cost-effective cars but reading the article I do not get the feeling that so-called SDV are in the interest of me, the consumer.
I think the article was focussing on the advantages it would bring to the manufacturer. Fewer control units, less wiring, hence a faster build time. Putting everything in one place is easier from a manufacturing point of view.
I guess it's the same as a 'bulletproof firewall'. Just a colloquial saying indicating both high importance and required quality expected for operation in strong adverserial environments.
As someone who has been working in security for past 10 years and systems / network admin for another 10 before that, I don't even know what a firewall is supposed to be any more.
Also, since I've worked on military systems a lot, I suppose a military grade firewall is just iptables for which someone has written a shitty gui (that might as well just be a webshell) and packaged it into a green rugged box.
Consider this. Almost every car on the road today has an unsecured bus going back to like the 1980s. However you need to actually access the car to do something malicious so the threat vector is zero; since if you have access to the car you can also just cut brakes or put in a pipe bomb.
The only reason why this paradigm changes in the EV era is because the insistence on having EVs phone home. Now you can concievably hack all EVs of this model at once and that is now realistic and even attractive to do. But again not a necessity for running a car. Just something that modern software focused companies want to see that leads to a host of expensive security issues that didn’t exist before. The car could be airgapped with the dealer network used to flash software updates like they do with most other cars before EV era.
The threat is not exactly zero. In some cases, thieves can get physical access to the bus from outside the car, and then inject messages to unlock it, start the engine, and drive away: https://kentindell.github.io/2023/04/03/can-injection/
Sure someone in that situation could also "just cut brakes or put in a pipe bomb" but car theft is a lot more common than assassination, at least where I live.
There are plenty of cars on the road today where theft is as easy as splicing two wires together. And yet grand theft auto isn’t very common at all even with all of these cars capable of being stolen in 10 seconds are being parked unsupervised on just about every block. Seems there are other filters in the overall system of society that are effective in keeping these unsecured cars from getting stolen today.
> Almost every car on the road today has an unsecured bus going back to like the 1980s. However you need to actually access the car to do something malicious
See [1] from 2023, where popping the headlight gives access to the bus. Lack of internal security then gives a way to steal the car.
The threat just isn't the same as the one you are modeling.
Security will come eventually, if only to prevent bad publicity.
It begs to ask why a headlight ought to have a data connection and not just power connection like most other cars of say 20 years ago. But even then when does the arms race end? Someone given enough time can cake apart a car to access any piece of it. A slim jim gets you to the hood release and the ecu of a say 2000 honda civic in 20 seconds. Was this a real world issue however in the 2000s, people hacking into drive by wire early obdii era cars like the s2000 to assassinate them with misdirected inputs or whatever the threat vector might be? Not really. Old fashioned ways to screw with people are simpler and cheaper.
I think anybody using this term has a shallow understanding of network security and just bundles it all mentally into a “thing” that stops all the bad stuff from happening.
Such a thing exists though usually not called “military-grade” per se. It is more similar to a data diode [0] than a classic firewall but has significant differences from either.
Data streams are converted into a sequence of objects that are required to have and satisfy certain formally verifiable properties as a pre-condition of forwarding. Any data or objects that cannot satisfy formal analysis requirements are dropped. Forwarding policies are only applied to objects that meet the prerequisite of being rigorously analyzable.
This behavior is bidirectional. It applies equally to data egress to mitigate internal threats and accidental data leakage. The internal mechanics can be pretty complicated and they necessarily operate on a store-and-forward basis. The data objects may be “laundered” by the firewall, what you send may not be exactly what the other side receives.
To make this work, the wire protocol, data representation, etc must be designed specifically to allow this kind of rigorous analysis and work well within these constraints. It usually won’t work on a random web stream and the data representation often sacrifices efficiency of storage for efficiency of verification and analysis at runtime.
In reality, virtually no one uses this type of tech outside of defense and intelligence because it won’t let almost any of the standard web stack slop through.
a firewall that prevents someone getting direct access to CAN bus and ECU, and sending messages like: "Key present", "Engine start", just by connecting to the wires of the headlight lamp (by prying a fender next to headlight)
I know that "military grade" has some relevant distinction in automotive. For example, normal car parts are designed to withstand "up to 80°C" and military grade means "up to 120°C". That has an impact on material choices and cooling.
So I have serious thoughts about driving “software defined vehicles” in the future. I mean, and the article has confirmed this sufficiently, the core competence of the established car manufacturers is not software. I don't trust the newcomers like Tesla or the Chinese manufacturers for the time being. In my opinion, the same standards should apply to software in motor vehicles as in the aviation industry. And there can't be things like permanent internet connectivity, on-the-fly updates or anything else that is suitable for consumer entertainment devices. So I'm seriously considering whether my next car should be an “analog” one - but it's going to be difficult, a Lada [1] (not so exotic in Germany, where I live) is only available second-hand because of the Russia sanctions. I'm happy to accept alternative suggestions!
Aviation standards are the way they are because if you have an engine problem you can’t pull over to the side of the road. But yes, something approximating these for road conditions is a good idea imo.
Part of me thinks the reason they are doing an integrated system is a combination of economics and convenience for 3 letter agencies to remotely assassinate ppl.
Having an engine problem on a back road is one thing, having a software-system-integration-what-the-hell problem on a Autobahn at 180 km/h +/- is a different story. And yes, I do not want my family in the car at that moment.
Projecting that "software had to be fully validated and finalized before the product entered production" was the stale old days and "make the car better over time" (i.e. out being driven) is the bright future by the automotive industry is far beyond worry.
Basically sitting inside a Windows that can kill you.
They all lost their minds putting stakes on software makers. I intentionally avoid the word engineering, engineering is far far away what is built up by the software making industry that is now tasked with being the babckbone of vechicles you put your and your family's life into. The cultures are incompatible.
(disregard mission critical software, their engineers are not proud members of the 'do not finalize, fix it later' bunch, not at all, they are nowhere here)
It states that consumer reports, (a for profit company providing independent reviews, and not a regulatory body) said the Model 3 stopping distance was not good. Allegedly due to a “bad ABS calibration”. Tesla released an OTA SW update.
Why wasn’t the bad calibration and degraded performance caught by regulators testing automobile safety standards?
The article also posits that this ability to make OTA updates expands the (IMO very very bad) SWE perspective that “it’s OK to ship unfinished and buggy products” into safety critical systems.
Ah interesting, I wonder if Tesla is an exception and if their systems do in fact follow ISO 26262. Standards are not necessarily legal requirements, and not necessarily checked by external people.
It sounds like their ABS system wasn't designed as carefully as conventional systems if there was such poor braking performance. Reading around, it might have been related to the emergency brake assist functionality not being calibrated properly.
The role of US regulators in the automotive industry is pretty different from what you seem to be expecting. They see their main goal is to set minimum, testable benchmarks for safety and give manufacturers freedom to achieve that in any reasonably justifiable way. The consequence of this is that almost nothing is required beyond meeting FMVSS and passing the tests it prescribes. ABS stopping distance is one of those tests, but a quick glance at the CR tests doesn't look like an FMVSS failure. The stopping distance simply wasn't up to industry norms.
Another consequence is that ISO-26262 and most other standards are completely, 100% norm-based in the US. They're used because the industry expects them, not because there's a legal requirement. You can deviate all you want and the only consequence is that regulators might take a closer look at your paperwork in the event of issues because they look unusual.
My car randomly braked today because it thought a car on a side road was pulling out. Not just sound the alarm but actually apply the brakes. Fortunately I didn't have a tailgater behind me.
I disable the "land assist" every time (which often tries to steer me into wildlife or other cars and was clearly not built for use on a single track country roads with hedges and random verges), but this was the first time in 3 years that the "front assist" caused problems.
If that's "high quality", I dread to think what low quality would be.
This happened to be on a highway when driving my friends car with all these assisted driving "features" while in cruise control. I was going up a small hill and for whatever reason there was a car stopped right at the top that I couldn't see. So the car slammed the breaks while I was in the middle of swerving out of the way. Which caused me to swerve more than I had intended. After I regained control it removed the breaks and attempted to return to the 80mph I was at previously which caused more problems because I wasn't ready for that.
I am now of the opinion that a car should never under any circumstance drive for you. If a car has cruise control it should cruise control you into a wall. That I can at least anticipate.
This happened to me a couple of years ago where the car I was driving decided that one of those water-filled tanks ahead of a barrier on a road under construction was in front of the car just because the road was curving hard to the right. It was very scary. It almost caused an accident by itself. I don't remember how the brake assist cleared, but the fact that there's nothing one can do to make the computer not break is very scary.
That happens decently often. This is the reality for all systems aside from braking system in trucks perhaps, which are more sophisticated.
The decision to do an emergency break is the same problem fully self-driving cars need. You need to interpret sensory input and have a model of the environment.
Ironically some genius made these systems mandatory despite them being a safety concern. Granted, they tend to work if someone really falls asleep behind the wheel.
This subthread is about ISO standards. The implementation quality of the new safety systems varies wildly.
On one of our cars, it is fine. On the other, it’s so bad they should have to buyback the last N years of vehicles. I hear the same high-variance story from friends.
The standards for this stuff are completely inadequate.
> You literally trust your life with medical devices full of software
I do not. A more charitable way to phrase that is "We are all expected to." And yes, well spotted, this problem extends well beyond vehicles. Or are you suggesting that this is somehow indicative that there are no problems? How would we all know if there _was_ an error in a device?
> those that conform to “piece of paper” standards, such as ISO 15708
That standard deals with non destructive testing and has no material that is related to the practice of medicine or the use of medical imaging scanners. It's not even the right piece of paper.
I don't understand the reasoning you are making here. Cars have had advanced safety features like emergency breaking for a long time now (e.g. Toyota Crown Majesta 2003). Furthermore, there are many safety features that are controlled by software (e.g. airbags and seatbelts) that exist in all cars manufactured today.
Some of them were caused by nothing other than faulty software. Which is the point. The more links in the chain the higher the overall failure rate. Which has implications on MTBF but also on overall failed deployments.
Just buy a car from the people, who dedicate their career/lives to making cars and have done so for decades. You aren't smarter then them. Your "serious thoughts" and "opinion" about what standards should apply are not yours to worry about.
Thank you, I'm fed enough with living in the world governed by the people who dedicate their carrer/lives to make it a peaceful, prosperous and free place (as in freedom) and have done so for centuries
BMW i3 is great for city/town if you’re OK with electric. Not 4x4, but minimal “assist”, just traction control. Internet remote stuff is optional but nice.
In the fully autonomous future the car I want to own and drive will still be my 6MT 911! :-)
Not to AUS or, as far as I’m aware, SFO. They also screwed the pooch in Austin by making Waymo available only through Uber, with no way to ensure you actually get a Waymo rather than a broken car driven by someone with a serious BO problem.
I think they can be separated though there's certainly overlap. But standards are going to be coming largely from pilots. Regulations are coming entirely from bureaucrats.
I’m not sure I understand everything you said but I went with Dacia Duster, it’s the affordable brand, but I like that I can have a new car that has the controls and everything like a car from a decade ago… (lol) physical buttons, relatively good quality as they get to rely on Renault’s everything, I don’t need to go to settings to open the glove box, they don’t try to “out-innovate” everybody with ads, subscription heating, goofy scroll-knobs, or non rectangle screens. You can put CarPlay and Android Auto in it if you want.
Also, you can just buy older cars, that works too.
BTW, I thought about buying a Lada Niva, because I love the looks, but I heard it is not that reliable as you would assume, and they are pretty pricey for a car that is basically the same for forty years…
They also have a poor safety rating from NCAP (at least they did 2 years ago), because they don't fit their cares with electronic aids such as emergency automatic braking, which is just another reason to buy one.
I visited Detroit last year and went to the GM headquarters. It’s open to the public with no appointment. You can wander around the Escherian maze with no guidance. A physical manifestation of every business decision GM has made in the last four decades.
That Niva is so nice! Just very very fuel inefficient, but man can it do off road in the hills of Albania. Take the one with the low gear and the diff-lock (and heated seats!). It's a joy to ride that thing (although not on the freeway). I also considered it, but even before sanctions is was very expensive due to taxes (here in western Europe). But it's so much fun.
I'd trust BYD more than Tesla but I don't want to have to trust anyone. I drove a 20 year old Honda still to this day, but literally every new car has software in it and it won't be an option in the future. It's just too profitable to gather the data that they generate. It's a privacy nightmare. I'm still appalled that Tesla got caught pulling footage of people having sex in their own vehicles, but the legal world has no intention of doing anything about it.
Also, as many of the well known manufacturers have been going for 40 to 60 years, and some of them for over 100 years (Rolls Royce, Ford, Mercedes, etc...) then 25 years is a newcomer :-)
As I understand it, yes the system worked as designed, but the design still managed to kill several hundred people.
I'm not qualified to evaluate the design of the system itself. Was it inherently flawed or would everything have been fine if the optional backup sensor had been mandatory, making this another example of corporate greed causing tragedy?
Either way, I don't think blaming the pilots is fair.
> a flight control feature the pilots didn't know about
phrasing that made me take issue. To me, it implies that if just the pilots had known about MCAS in detail, the crashes wouldn't have happened, implicitly placing the blame on them. While the reason for both the faulty system and the lack of knowledge can be attributed to Boeing trying to save money, which I think is a more fair target for blame.
But that's the point! A professional pilot misunderstood/was unaware of a new safety feature, despite their professional experience and continuous training.
So, is it really sane to put similar features in cars, where you get your driving licence at 16/18, and then that's it?
This also goes for the huge screens on the console. A pilot has been trained for each commercial aircraft model they fly to navigate their way around the numerous controls. But putting a tablet in front of an untrained driver? It sells well because it makes you feel as a pilot. But at the same time, it is a huge distraction and there is zero training to cope with it.
It was Boeing that intentionally hid the importance of a system much more relevant than flight characteristics of a plane. That is an intentional violation of the spirit of the safety checks.
Terrible mobile website for what its worth. Two sentences per in paragraph ad and I couldn’t fully read the article because it bogged my se2 down to a crawl. How I wish I could jailbreak this phone and install a real adblocker but alas not on magic version number.
> Access InsideEVs and over 450 other websites as e.g. motorsport-total.com and formel1.de without banner ads, personalized tracking and video ads for only 3,99 € per month.
If you want to know the many ways this is going to suck, then think about everything you've ever heard someone bitching about in the modern video game ecosystem, then multiply it by "but instead of people not being able to play a video game, someone might die".
Is this how we get the Butlerian Jihad? Because part of me sure does want to learn how to identify cars built like this and learn ways to disable them when I see them parked somewhere around town, before one of them fails to recognize me on my bicycle as something that should be avoided.
Give me a car that is perfectly 100% autonomous, or give me a car with three gauges and basic controls only. Everything else is an uncanny valley: all the downsides of complex tech without being useful enough to justify it.
Until then I like my Nissan Leaf: physical controls, phone just docks with infotainment screen, and reliable.
After using it for 3+ years, I'd really miss automatic cruise control.
You can an intuition pretty quickly for what it does and what it doesn't, and in certain situations it really takes a lot of attention off your plate (stop-and-go traffic, and long distances on the highway).
Remove the LTE chip and all functionality related to ads, support wireless CarPlay and android auto, and use physical buttons. You’ll win every award in the industry.
Yeah all I want is something that holds the phone and gives it a USB-C port that charges it, lets me play media through it, and lets buttons on the wheel control the phone (volume, next/previous, and programmable go forward/back x seconds buttons).
USB-C is so powerful, it can do everything Bluetooth does while charging, but for some reason that's just not an option in a lot of cars? Make it make sense.
It's the latency that kills me. Let alone the stupid "you're an idiot if you let this screen distract and kill you" message that pops up, it seems to take a good 10+ seconds to sync with the phone and "come alive".
This is with USB, too.
I want the car to start and CarPlay to be operational; we have no time to be wasting on whatever formalities software wants to have.
Maybe someday wireless CarPlay could start syncing with the system before you even get to the car, so it's already loaded when you sit down and start.
But some do, don't they? It seems to me that Hyundai will initiate the phone connection right when you open the driver's door. Then, as you sit down and start the car, the infotainment has already booted up and the phone connection comes online almost immediately.
Also, during short stops, the screens go black but the connection is kept up, so when you re-start, there is no delay.
The Slate truck won’t have CarPlay because it won’t have infotainment at all. If you want Carplay buy one of the dozens of offerings from companies like Alpine and have it installed at one of the thousands of stereo shops across the country. You’ll get exactly what you want from experts in what you want.
I mean I have that already, a Volkswagen E-Up that has a cradle for a phone with a USB port behind it for charging. They even have an app that connects to your car directly(through Bluetooth! No fancy subscription based nonsense) and shows you all charging/energy consumption figures.
I just mean I'd totally buy a much higher end car that is like this, I don't need a screen with all the nonsense on it.
My car even has a relatively small console screen but still prefer it over my (non-plus size) iPhone. I could live with just my iPhone on USB but consider the center screen a plus. (The vehicle is pretty good about climate control etc. on buttons.)
My old ford tourneo custom has a well placed phone holder. I use this and a MagSafe charger plugged into usb port on dash. Works great, I can use my phone, or anybody else can use the van and their phone, it’s really easy. Just looked and newer models have great big touch screen instead :(
Physical controls are worth it for me. Having a press to talk button, track advance and volume controls on my steering wheel is a pretty nice quality of life feature. I could do without a screen if the car has that.
You get a new holder. The one that came with our e-Up was too small for my S24 Ultra, so I just got an adapter on eBay for like £3 and installed a new holder(with wireless charging!). Where there's a market need someone will provide a solution.
I'm not sure why you have the downvotes. Even it's mostly just about GPS, the built-in screen is better than iPhone on a somewhat dodgy clip attached to a vent someplace. Unless the car were otherwise compelling--and it's a pretty competitive market--not sure I'd buy a car without CarPlay.
The more premium Opel Astra K models had a phone holder built-in (optional accessory, but you get the point) that would also charge the phone. Practically no one bought it so they removed it in later models.
Does anyone prefer wired CarPlay over Wireless CarPlay?
I was annoyed enough that our used/new-to-us 2020 vehicle only supported wired that I bought a wired-to-wireless adapter and brought it with me on test drives to ensure that whatever I bought would work well in wireless mode [or else I was buying a different car].
I installed a wireless charger under one of the cubbies that was well sized to hold my phone on long drives. No need to faff around with cables.
> Does anyone prefer wired CarPlay over Wireless CarPlay?
Yes, for the main reason that I have a Starlink Mini on my roof rack.
My phone can connect to the vehicle via wifi, or it can connect to the internet over Starlink via wifi, but not both simultaneously. With wired CarPlay, that problem is solved.
That's just for cables, I think you still need MFI (and all the paperwork and approvals and cost and the chip itself) that go with it if you want to implement a CarPlay receiver.
You can't, it's required for eCall which is a mandatory feature in Europe.
Unfortunately, it's fraught with issues, especially for the very first eCall modules where the hardware supported only 3G (HSPA)... which is being phased out across Europe together with GPRS (1G)/EDGE (2G), leaving these cars without a working eCall system - and no upgraded hardware modules in many cases.
I mean we can also change laws again in europe (in favour of that) - but we could also keep it as a separate module. So the LTE chip only gets used for an emergency call and nothing else. No remote control.
Unlikely to happen, but possible (not 100% safe, but good enough).
Was my sentence that unclear? I constructed it as a OR. Either change the laws (my favourite solution as I don't like enforced modification of my car) - or use a technical solution to just meet the law.
If you don't turn on the LTE chip, it can take a while to get registered or a network in order to perform an emergency call. If the LTE module is turned on (the same applies for mobile phones) the network operator will know your coarse location because of the LTE specification itself. Furthermore, all communication equipment is legally required to support Lawful Interception; and LTE is no exception to this.
And naturally of course you don't carry a cellphone with you while you drive...you know, that device with accelerometers, GPS and an LTE chip that you leave powered on all the time on your person?
You are also free not to use a mode of transportation. You can, for example, buy an old car, register it on a historical number plate, and you’re good. There is no retrofit requirement. If you want a new car, suck it up.
I'm always amused at people who deem future and freedom of choice incompatible and effectively stating that the end game of all progress is absolute slavery and obedience.
Other than having some vested interest in denying people's right to choose between positive and negative aspects of progress, or having a Stockholm syndrome towards those who have such interest, I got no plausible explanations to this psychological phenomenon
Ah yes, brilliant reasoning. "You carry one device that can be tracked as long as you're carrying it so what's the problem with having tracking baked into your vehicle?"
I can actually remove my number plate. One tool, five minutes, car still drives.
And of course it's a tracker. It reports my location to a third party. There is no other definition for it. That it purportedly only does this during an "emergency" is not something I can verify nor trust.
I’m certain you aren’t that important for anyone to care.
Yes: it reports your location to a third party. When you have an accident. So you can get help.
And of course it can be verified. At the end of the day it’s a sim card in a passive mode. Maybe you don’t know how to verify it. There are millions of them driving around. I’m sure someone more qualified than you would have already reported if they were tracking in real-time. The system is nearly a decade old.
> I’m certain you aren’t that important for anyone to care.
Storage is cheap. We know the NSA hoovers up everything it can and puts in a database. This is facile reasoning.
> So you can get help.
How many accidents resulted in fatalities because 911 couldn't find the victim? What's the percentage on that you think? Is there _no other way_ to get help other than this system?
> I’m sure someone more qualified than you would have already reported if
You're just full of confidence. Pray tell where it comes from. What makes you believe this is true?
> You're just full of confidence. Pray tell where it comes from. What makes you believe this is true?
I trust in people who look for this stuff as a daily job, or because „they’re here to get you”. Because you know a sim card in the open is really easy to verify. And there are thousands of them driving around but also easily approachable sitting targets in every possible setting.
But you double down. I find this discussion pointless.
> How many accidents resulted in fatalities because 911 couldn't find the victim?
You’re missing the point of the eCall system. Think out of the box. What if there is no witness? What if everyone in the car cannot get help
Everyone who could reach out for help, received it. Can you tell the same about those who weren’t able to contact first responders in time? Maybe they’re still missing! It’s called survivorship bias.
Can you explain why these protections are not sufficient for privacy?
> 112 eCall is not a black box. It does not record constantly the position of the vehicle, it records only a few data to determine the position and direction of the vehicle just before the crash and these data are only transmitted to emergency call centers if there is a serious crash.
> eCall cannot be used to monitor motorist's moves. The SIM-card used to transmit the eCall data is dormant, i.e. it is only activated in case the vehicle has a serious accident (e.g. the airbag is activated).
Well, for starters, the SIM card in the eCall system is only needed to receive a callback from the 112 service if the call drops, which the system automatically picks up. You can actually dial 112 without a SIM card in your phone, as a GSM device without one can still connect to nearby radio towers in a "limited-service set," better known as "emergency calls only."
I know some carriers have strange quirks in their SIM provisioning systems. For instance, it may take more than a few minutes or require a specific type of coverage (like UMTS) for an otherwise "dormant" SIM card to activate from limited service on short notice.
>112 eCall is not a black box. It does not record constantly the position of the vehicle, it records only a few data to determine the position and direction of the vehicle just before the crash and these data are only transmitted to emergency call centers if there is a serious crash.
That statement is factually inconsistent. Either 112 eCall incorporates a time travel device or it must constantly record the position and direction of the vehicle and other data. In theory, that data is then deleted, but you have no way to verify that it is - and it would only require a trivial, unnoticeable software update to modify this.
Thankfully, we're safe. Car software is notoriously high quality and rarely hacked. All governments are fully trustworthy, especially around espionage and privacy, and have a perfect track record of never lying to the public.
Look, the European Commission stated that it cannot be hacked; "hackers cannot take control of it", from ec.europa.eu. They built an unhackable device. I am not sure what you could be worried about. If the government tells you something cannot be hacked, then it cannot be hacked. Furthermore, none of the EU member states have been found using other infrastructure to violate privacy laws.
It's not. It just stores the last speed/wheel position/brake state data that it receives when the "collision imminent" condition activates. In some cars this can be literally the same signal that deploys the airbags.
> Look, the European Commission stated that it cannot be hacked
Pretty much. It's just a normal LTE radio, that is normally inactive. It technically is hackable, but I'm not aware of any hacks of baseband firmware of this severity.
Sorry, that's incorrect. I have actually read the law and its relevant standards. The standard requires at least two pre-accident locations to increase accuracy and other fields with pre-crash data are encouraged.
And come on. Car manufacturers, which are notorious producers of insecure software, are legally mandated to make an inexpensive device which includes an LTE radio and a connection to the vehicle buses, and you think that is... unhackable? I can't tell if you're trolling me, but your average blackhat only needs 1 of (shitty car OEM software/LTE radio/vehicle bus connected device) to break into a system. This system is a trifecta of hackable crap. To call that, of all devices, "unhackable" is priceless.
The MSD (minimum set of data) is defined in: "CEN 15722 ESafety - ECall - Minimum Data Set".
The original standard version defined only one location datapoint, the more recent version defines two additional _optional_ points ("recentVehicleLocationN1", "recentVehicleLocationN2"). It also allows specifying the number of passengers.
The mandatory datapoints include the location and direction of the vehicle, but they can be acquired as needed.
> I can't tell if you're trolling me, but your average blackhat only needs 1 of (shitty car OEM software/LTE radio/vehicle bus connected device) to break into a system.
I'm not aware of black hats hacking into a modem that is passively tracking the mobile networks. It's theoretically possible, but I'm not aware of such feats.
Sorry, your comment is incorrect again. The most recent version of CEN 15722 requires the two most recent locations before the incident location.
The modem does not have to passively track the mobile networks; it can do what it wants. The common OEM implementation these days is that the physical device that does eCall does several things, including eCall, over the same cellular radio. There's nothing stopping the OEM from connecting to a random website and eval()ing the result.
You seem confident in the security of this unhackable system so I will point out some of its other security weaknesses. Several eCall device implementations include Bluetooth modules (both "unused" as part of hardware and implementations that use Bluetooth.) Bluetooth is as secure as a wet cardboard box, so you could take BlueBorne or one of the six million other Bluetooth exploits that work on a non-discoverable device just sitting on the shelf, get in that way and boom, you can transmit whatever you want over the cellular radios. Vehicle infotainment systems are pretty insecure on average and are frequently hacked, so you could take over the infotainment system, get into the CAN bus that way and then send bad data to the eCall system, which is in the business of processing and responding to CAN data.
But those are just a few of the million ways; you could write up attacks all day long and you wouldn't scratch the surface. The facts are: this is a system with cellular radio(s), a CAN bus connection, sensors that constantly listen and interpret data; this is a large attack surface, built by OEMs that write notoriously insecure software. It is, by any reasonable judgment of those facts, a pretty hackable system. And yet, the European Commission goes around telling people that it "cannot" be hacked.
Anyway, this will be my third comment in a row here telling people that their comment is plain incorrect, so I'm going to have to leave the discussion here. I hope that my words provide some food for thought - for the next time that a system that could track you becomes legally mandatory.
> Sorry, your comment is incorrect again. The most recent version of CEN 15722 requires the two most recent locations before the incident location.
I have the standard open, and I don't see it.
> The modem does not have to passively track the mobile networks; it can do what it wants.
Sure. So just choose an automaker that doesn't provide data subscription services and/or don't pay for them. The eCall requirement in itself doesn't require tracking.
I can't believe I'm still responding to this, but your comment is incorrect. Again. If you are reading a version of CEN 15722 where the two most recent locations before the incident location are not required, then you are reading an outdated version from over a decade ago, which has been withdrawn in favor of a more recent version for some years now. The year is displayed prominently on every page of the standard, and the validity is _required_ to be indicated before you even open it, so I have some doubts if this is a genuine mistake.
It's been a while since I've seen the pro-surveillance argument of "well, you can just do this uncommon or difficult thing if you want to evade surveillance!" In several European markets, for several vehicle types, there _are_ no vehicles without OEM-connected cellular radios. Some OEMs don't even advertise it; it is used for activating features, or "security." Other OEMs will not _fully_ disable cellular connectivity even if you stop paying. If you're an OEM mandated by EU law to include a cellular modem and a location recording device, you might as well make some use of that - and they do.
Your privacy should not rely on the government's "trust me bro" and it's like we forgot about China owning SS7 last year lol. No, it is plainly obvious to me that the mandate for eCall and the lack of an owner off switch for it is for nannying and surveillance FULL STOP. No option to opt out must be treated the same as the violations to privacy they are attempting to pre-construct the conditions for.
Almost like sometimes people get seriously injured in car accidents and can't get to their phone, assuming it's where it was left prior to the accident.
Let people decide if their fear of death warrants all kinds of surveillance on them and not shove that down their throats as in "we know that many of you would opt-out of being tracked so we made it mandatory", alright?
There's nothing at all sensible about legally mandating that all cars must have a device which can be used for surveillance. Still less so for something which is at best a marginal gain in safety.
This is the same way that hn proclaims every single arxiv paper as revolutionary. I really wonder sometimes who is this gullible on the internet (kids? bots? I
influencers?)
I don't even think they've built a single prototype. I'd be happy to be corrected but last time I checked, none of the "prototype" shells they showed off had a powertrain.
I mean, they did something, for sure, but they sure as hell didn't do "this" ;P. What they are doing is more in the line of not providing even hardware, much less software, which is an entirely different paradigm... like, they don't even provide speakers?!...
All depends on how they market it. Wind down windows to me today is an aesthetic statement - “we are selling a cheap, no frills vehicle - look see! Even wind down windows”
Such positioning could be what the intended customer base react well to.
My biggest concern is lack of a stereo. Did they include speaker cutouts and wires, or are you looking at a $1000 labor bill, minimum?
I’d much rather they included a $200 system, since ~ 100% of their customers will want to be able to have speakers in the doors and a mic in the dash (at the very least).
I seem to remember Jeep saying manual window winders were actually more expensive once you factor in the costs of having them as an option given how cheap electric ones are when they dropped them for the new Wrangler. Might still be cheaper if you only manufacture with them and don’t offer electric but the price difference can’t be that high.
I never had a manual window winder fail to work, but electric window buttons breaking or the motor getting stuck (e.g. in icy conditions) has happened at some point in every car I've owned.
The convenience factor hugely outweighs the rare failures for me, but I could see why someone buying a Wrangler for its intended purpose might actually prefer the manual option.
Electric winders tend to experience motor failure leaving you with a stuck window. They don't stress the mechanism that much.
Manual winders, you would generally experience mechanical failure. If you were lucky, you broke the handle. If you were unlucky, you broke the gears and had the pleasure of watching your window fall into the door.
I think the market for base Mavericks is pretty small. At that point it's really not providing much value over an SUV with rear seats that fold down. I agree not everyone wants to spend 40-80k, but that doesn't mean they want to spend $20k for a small no-frills EV in the shape of a truck with not many other similarities.
I like the "starts out cheap, then upgrade it later" premise of Slate, and I like that it's electric, but it'll only really be a toy with the limitations I specified.
AWD is only standard on the fanciest Maverick trim and not an option picked by the couple of folks I know with them. But that + the bed length doesn't seem to be stopping them from loving their trucks. Tacos also start at 5' IIRC.
But if you have even just those once-a-year "need a truck bed" needs the gap between "SUV with fold down seats" and "actual truck" is pretty substantial.
I think the set of truck buyers with either:
* just occasional needs for a bed, without a need to put sheet goods flat or such (if you have that just get a minivan these days ;) )
* a fashion-driven desire compared to a van or SUV vs a practical-driven one
is substantial compared to the set of "needs a professional-grade truck" buyers.
The set of professional-grade buyers hasn't changed much in thirty or forty years, but the former two sets have exploded.
For a pickup truck, if you're buying it so it's "there when you need it," it removes a lot of value if it's gotta stay in the garage during the winter.
Though yes, I could see it being less important if that's not an issue in your area.
It's not a truck. It's for people who need more than a car, but less than a truck. A Cuck? (oh, wait...) A Truar? Either way, those specs are plenty for the average person that just wants to haul some stuff, or pull a small trailer, and not burn too much energy while doing it (or not doing it).
Mazda has done a great job at this so far, very minimal screen which automatically just shows CarPlay, and buttons for all the normal car stuff, which also isn’t overdone. The only flaw is the scroll wheel to interact with the screen, which is just slightly too clunky in apps with too many options
Mazda also managed to squander a huge brand and structural advantage by falling into lockstep behind other Japanese automakers in underinvesting in EV manufacturing infrastructure. Now they have to rely on their JV partner Changan to lead the way in producing EVs, giving up the core structural strengths that Mazda previously had in designing and building their own components - including software and controls, which in the Changan-led models have no continuity at all with Mazda's domestic models. They just superficially copy the Mazda exterior design language while wholly dependent on Chinese supply chains (and some Android Auto for the software, it seems) for manufacturing the actual EV.
A lot of people still don’t want or can’t really afford EVs given their limitations. I’d say it’s the majority where I live. I directly know only one person who has a full EV (not a hybrid).
I don’t think the Japanese automakers have squandered anything, yet.
We paid maybe a $10K premium for a used EV truck. It gets 2mi/kWh. Most parts of the country are paying ~ $0.125 per kWh, so that’s &0.06 of a dollar in electricity per mile.
A comparable truck gets 18mpg mixed. At $3/gallon, that’s $0.16 per mile. So, the price premium pays back after 100K miles. That’s comparable to milage driven during a long car loan.
I ignored oil changes, tax breaks on used cars, and picked the form factor where EVs are the least economical.
For commuter with charging access at office or home EV makes sense. For me making 300+ mile round trips with no charging infra (pull in at the gas station in the foothills) and low overall mileage EV is trickier.
There's a stop, it's just in the middle of nowhere. 20 minute charge would be annoying but survivable assuming they had them in say Lone Pine. And yeah, 20 something mpg and 16 gallon tank multiplies out to a large range.
Including supply and generation, we pay $0.148 / kWh, and yeah, I average $0.06 charging my EV at home on a slow, inefficient 120V / 15A. (Some day I'll upgrade, maybe.) I've never charged anywhere else (except for free at the used car dealer where I bought it.)
We make a ~180 mile trip roughly once / month and could charge on site as we always stay ~2 nights, though probably slow 120V / 15A charge (aka Level 1). My current car would probably be pushing it, range-wise, but I definitely think for the vast majority of our usage, we could be using only EVs if we got one with a 300+ mile range (based on 100% battery usage.) From what I've read some EVs (like mine) struggle a bit below 15% and start to run in "limp" mode.
China is currently making affordable EVs, though they might not meet (American) expectations of things like range. There’s no reason why traditional automakers couldnt be doing the same had they not focused on larger ice vehicles, hydrogen, or the luxury market.
Hopefully they figure it out because I love my Mazda 3 hatchback and would buy an EV version of it in a heartbeat. Not only is it very fun to drive (I have a manual transmission) but the interior design is excellent.
Mazda's target market is quite different from the EV buyers one, at least here in Europe.
Its reputation is that of a brand for people who really like cars, who can appreciate the care put into proper engineering and a wonderful manual transmission; or people with an eye for a "conservative" kind of quality. It's basically the new Volvo, but sportier.
While that might affect their market share in HN neighborhoods I assure you Mazda is making money hand over fist selling their boring non-hybrid SUVs to normal people. People love them and they sell.
It's not about unusual, just that a low profit margin in a volatile industry is (with a downward trend in sales for almost 10 years), by definition, not making money hand over fist. That is why their market cap graph looks like this:
Mazda’s operating margin is higher than Walmart’s (along with many others). I think hyper scalable sectors like high tech and finance distort our OM expectations.
Operating margin is irrelevant, only profit margin matters for this context. Walmart hangs out in the 2.5% to 3.5% range, not materially different than Mazda. Either way, any business with a low single digit profit margin is not making money hand over fist. It might be different if Mazda had such a huge and loyal market share that their low profit margins are offset by low volatility of expected future sales (such as with Walmart/Costco), but that isn't the case at all with Mazda.
Their expectation is that their sales will be stagnant at best, but probably decline for the foreseeable future.
I know Mazda makes good boring SUVs, I own one. I like Mazda's design philosophy, that's why I want them to succeed. In terms of vehicles sold, Mazda's sales peaked in 2017, the year before I bought my most recent one. As best I can tell, operating profit peaked in 2016.
Mazda maintained their relevance and independence by operating their own center of design, engineering, and manufacturing excellence in Hiroshima, and exporting the results to the rest of the world, since at least the 1960s. As I mentioned, that thread is now broken as far as EVs go, with the Changan JV making EVs for Mazda. China is now producing excellent EVs that surpass the capabilities of ICE cars at a fraction of the cost/price, thanks to continuous improvements in LFP battery technology. China also dominates solar, which (together with the batteries) solves the grid stress issue for large EV deployments in most regions of the world. Together these exports are likely to disrupt Japanese, US, and European ICE exports and energy markets throughout the world, no matter what tariffs the US chooses to enact.
Mazda and the rest of Japanese companies slept on it, led by Toyota's trust in the hydrogen-powered future that didn't materialize, even while Panasonic had the best batteries in the world. The time to invest in these platforms and technologies was 15 years ago - now they will have a far harder time financing this and finding technology development partners. Sure, they can survive - not thrive - on existing ICE exports for a while, but they will face a shrinking market and stronger headwinds - and are likely to lose their independence, which is what allowed them to design great cars. Don't believe me? Look into what's going on with Nissan (which squandered an even bigger lead - the world's first mass-produced EV).
My 2017 Mazda cx5 refuses to not play the radio. There is no "off" for the audio, you have to choose a source. I use my phone, via bluetooth. But sometimes, for unknown reasons, the car does not connect with the phone. It then falls back to the last source chosen before BT, which is radio.
Okay, so I created a flash drive with an mp3 of 30 seconds of silence, played that, then went back to bluetooth. This failback strategy worked one time, then it also failed to recognize the flash drive, and failed back to radio, again.
I will never want to listen to the radio. I would love to remove radio as an option. I would love to have no fallback as an option. But no, the car just f-n loves the radio and will not stop trying to force it on me.
This car definitely tries too hard to be smarter than it is. There's all sorts of exceptions that keep the doors from auto-locking when I walk away, and I would turn all of them off, but I can't. Walk away too fast? doesn't lock. Open the rear? won't auto lock. Car just doesn't feel like it? doesn't auto-lock.
And god forbid you hit the unlock button when the passenger has already unlocked it. Anxious beeps from the car for several solid seconds. That is not an error condition!
Performance and reliability have been great though. They just need to stop trying to be smart. They're not.
Long pressing the source button turns off audio and keeps it from turning on automatically on the next start. This at least lets you explicitly decide when you want music.
While not dramatically better, just a few posts down[0] someone paid for the "Welcome to Mazda" service manuals/program for $30 and shared the fuse box schematics
I have tried that (not on a Mazda). The radio is still there playing whatever and if there is a valid station the now playing song has to be shown on the other useful screen. On I got the system to default to radio off, but that means I can't control my heated seats w=ithout turning the system on - there are several seconds of noise between getting the system on and it responding tol the volumn knob.
I think making manufacturers pay you back the whole car in a recall, or half the car and you keep it, for this kind of crappy design, would be a good thing (especially since I am sure the firmware is code signed lolol). Oh no more Matsuda or GM because they went bankrupt from fines and restitution? Cry me a river, sucks to suck cutting corners lol.
Re: radio always turning on, my LDV eDeliver 9 is the same but worse - sometimes the radio comes on immediately, and sometimes it takes about 20 seconds. You can’t preemptively mute it in the latter case. There’s lots of other weird quirks with the radio (e.g. going into reverse switches to a low-volume radio if you were previously playing music or a podcast in CarPlay). It’s as-if almost any change in the audio switches the radio on.
Other than that, it’s a great van!
I bought a Mazda3 a few months ago and I love it. It is exactly what I want as a driver.
I even adore the scroll wheel and wish it could be in any car I own in future. Yeah it takes slightly longer to do certain actions in CarPlay, but I can do it so much more safely than I could in the Civic I had before. The infotainment boots basically instantly; as you mentioned CarPlay starts itself, and the patronising-but-mandated “don’t use this in motion” warning dismisses itself. In the Civic I would be half way down the road already by the time it booted, blindly prodding at the screen to try to dismiss that warning so I could pause the podcast that started playing itself because I plugged my phone in.
And, while my 2022 car predates the stupid auto-re-enabling ADAS requirement in Europe, the 2024+ models have single button deactivation. I dunno how, cause it’s supposed to require a minimum of two presses legally, but it sure makes me wanna stick with Mazda.
However that makes the upcoming 6E that much more disappointing. They’ve partnered with a Chinese manufacturer, I assume because they don’t have an EV platform of their own ready yet. Looks fantastic from the outside, but the inside is a sea of touch screens with barely a physical control in sight.
Generally agree but they are laying the path to enshitification. You see you can get turn by turn directions on the HUD, but only through their app where they want you to pay $10/mo for the privilege. Same for inputting addresses into their crappy nav system.
So I only use Google maps with Android Auto now, but cannot put the turn by turn display on. Also, who knows what telemetry Mazda is sending home on me without me knowing or wanting them to. Probably selling it to data brokers.
I believe I've heard the newer Mazda 3s have added the navigation into the HUD for Android Auto and Carplay. It's not in my 2020 though which is annoying.
As for selling your data, yes absolutely. It goes to Connected Analytic Services which is an affiliate company of Toyota Insurance. Toyota Insurance Management Solutions (TIMS) is another name to look up. Subaru sells your data to them as well.
When I was doing my car shopping two years ago, I was initially considering another Mazda, specifically looking at the Mazda 3 AWD Hatchback. Their high tech features were significantly behind the other Japanese auto manufacturers. Some features like the ability for the car to automatically stay in a lane were not present.
When looking at who is doing it right, I wouldn’t put Mazda on a pedestal. They simply are behind the competition.
That’s interesting - what vehicles require you to do that? I know the usual suspect is the Tesla, which I have, but I never have to navigate through menus to change the temperature while driving.
As an aside a lot of people like to levy criticism on the infotainment screens which I think is very well deserved, but then people text and drive, watch YouTube videos, and do all sorts of crazy things too.
Instead of levying criticism on these distractions (let’s include billboard too) we should instead focus on just reducing car usage since we won’t stop people from being distracted.
> As an aside a lot of people like to levy criticism on the infotainment screens which I think is very well deserved, but then people text and drive, watch YouTube videos, and do all sorts of crazy things too.
Instead of levying criticism on these distractions (let’s include billboard too) we should instead focus on just reducing car usage since we won’t stop people from being distracted.
This argument to me reads like one for abstinence from sex. The world is not so binary, we can both criticize distractions and build communities where car use is not a necessity. Not to mention in most jurisdictions some of these distractions are criminalized.
We can - but we don’t need to clutch our pearls about infotainment screens as if they are some sort of special moral insult relative to what’s very common in today’s driving communities.
Criminalization of texting and driving and such doesn’t matter unless you enforce, and we don’t enforce. So it’s de facto legal. Who cares about infotainment screens at that point?
We are talking millions of cars driving at a speed that can kill people both inside or outside the car. Anything you can do to reduce those distractions is a net positive for society. Less death.
As for criminalizing texting, I’ve heard enough people getting caught and getting big fines that it works enough for me to dissuade me from doing it.
If you live in America it’s just not enforced. Even cops do it. I don’t do it because I just don’t need to but you can watch people doing it for yourself if you pay attention.
If you’re focused on less death, sure we can criticize infotainment screens, but the energy is much better spent in demanding enforcement and in whatever we need to do to reduce car usage. Otherwise you’re kind of wasting your time, unfortunately.
I've got a Ranger with SYNC 4. HVAC has its own set of buttons and dials below the display, but you can use the touchscreen widgets too if you want to.
As a Linux fan and owner of a Sync 1.0 vehicle I feel your pain. I want to replace it with something aftermarket, but the cost of a dash kit is pretty steep if you want one of decent quality. I reboot it weekly, which takes minutes, so it doesn't freeze during the week. I'm guessing there's memory leak that takes a while to accumulate.
Just bought a 2022 Mustang to avoid the 2024-2025 series for SYNC 4 and the removal of physical buttons. Car is nice but can’t get past the whole digital set up.
Subaru require you navigating to second screen for climate modes. Simple temp adjustment has buttons, but the screen interactions for basic usage feels dangerous as a driver.
I’m being pedantic but the OP did specifically say they need to “navigate through screens to adjust the temperature” which I think is different than setting climate modes. Not that I’m defending that you might have to do that specifically, but I was responding to the OP’s specific wording.
We rented a BMW which had all climate settings on a touch screen. That touch screen crashed once and we couldn't turn the air con off without trying to reboot the car which isn't exactly trivial since there isn't any obvious off button.
The Nissan Leaf is (was?) what you describe, apart from the LTE chip. The LTE doesn't seem to do much without NissanConnect (which was actually written by Bosch).
Nissan tracks you and sells your data. Pretty much every manufacturer does, if your car has a modem rest assured that your car is collecting and selling data from you.
Removing LTE would remove key features that drivers want, including real time traffic updates, remote controls and streaming media? What's your objection to LTE?
The moment you pick a non-techie off the street and help them see the amount of data collection occurring, you have another person who proves the assumption. It's not a huge assumption.
No one likes ads, no one likes their data being collected. The sooner insurance and car companies understand that, the sooner they get out of the maelstrom of false revenue from ad- and spy-ware programs.
I don't think it's a huge assumption. It was in the past, but not anymore.
The thing is that car manufacturers have been fucking up software in cars since... forever. The second car play and android auto hit the scene, that's all anyone wanted.
There's more benefits than just what's on the surface, too. Even if the car software is perfect, it doesn't have access to the same data your phone does. It won't put your contacts in your navigation, for instance.
Right, and Apple Car Play does it out the gate. So much so that I can say "Navigate to Doctor X" and it does it. And it did it without convoluted requirements on the vehicle side. And it will continue to do it, because Apple's navigation isn't going to rot like the car manufacturers will.
Look, can car makers make somewhat decent software? Probably, if they burn enough money. But is it even worth it? I don't think so. People already use their phone hours a day, just let them use that.
The only problem with CarPlay (and presumably AA) is lack of integration with the car…
Changing lock, light, and anudio (bass/treble/sub/fade) options. Map integration with fuel capacity (they only recently do this for EVs). Checking service intervals, recalls, etc.
If CarPlay had APIs/toolkit to serve those functions, it could 100% replace the UI that the manufacturer delivers (and nobody likes).
That's a concern for privacy focused individuals, who are a very small fraction of the consumer market, despite being common here on HN. If the last few decades have shown anything, it's that most consumers don't rank privacy highly as a desired feature for products in anything but the most abstract ways.
There's zero chance a car manufacturer is going to nuke some of the most desired features of modern automobiles for some undefined cohort of privacy conscious consumers.
Most younger drivers would even buy Chinese vehicles despite their privacy concerns.
I just want to be able to disable things even if they're default on. Or if not a software toggle, perhaps pull out the SIM card so the connectivity goes away.
Some privacy concious customers know that data collected by China is less relevant than data your own legislation could get access to, but that is a different point.
It just should be said, that all these features could perfectly be implemented without violating privacy. You just have to use another system not from Apple and that other advertising company.
I have never once seen someone use the manufacturer provided traffic data, navigation, or "streaming media" over their phone when given the choice. Let's be real; it's just an excuse to try to subject customers to another subscription fee.
Removing LTE doesn't cost me real-time traffic updates because (preferred maps app) is running on my phone which already has LTE. Streaming media? The media is being played from my phone or streamed via my phone, which already has LTE. I'm not sure what "remote controls" are in this context? Letting me set the A/C fan to high from Internet (almost certainly via a browser or app running on... wait for it... my phone)?
We've already paid for the LTE modems and app integration on the phone side of things, don't need to pay for it a second time on the car side or have to deal with the vehicle manufacturer's terrible implementations of navigation apps and media streaming services or yet another vendor collecting telemetry about me and reselling it to whoever wants to pay.
Why does anyone need any of those except maybe remote start. The rest are handled though CarPlay. Nobody wants built in navigation that the phone already does
Don't know about the rest of the world, but the EU requires e-call (automatic emergency call after an accident) for all new cars now so you can't sell cars without an LTE chip.
... but you can be a bro and make sure that hardware is close to the surface somewhere for easy access, its presence isn't required to start and operate the car (either firmware check or the immobilizer performing metrics), and its removal does not cause an obvious and annoying alert during operation (IE removal should not make the car appear to be in a 'degraded' state per its indicators).
You are complying by installing it, the customers are the ones [easily] removing it [because you were a bro].
It is mandatory to have and it is in checklist during annual vehicle check. Without it it is not street legal. And the car should show an error in case the module is removed/failed.
Yes, that is correct. But it is a different concept. The mobile network operators do that. Not the car. The car does not even have any way to stop that apart from going offline.
First of all, you didn't directly answer my question, it wasn't a yes/no question. I don't know how these systems work. If they're almost always offline it would be a lot less concerning.
Secondly, the government makes both regulations (cell tower data retention, emergency calling). So in combination the government regulated user tracking. Where exactly it happens isn't really that important. Once the data is being collected it'll end up being used. Usually it starts with major crimes and anonymized statistics, once the automated infrastructure has been created it'll be abused either illegally by police overreach or legally by a tough-on-X government changing the rules.
It's not like this is some novel pattern...
Really, it's only surveillance, thinly veiled as safety. The reality is that my car is no safer with an LTE modem. Zilch. Zero. It's purely surveillance, if you don't fall for the marketing.
In fact, I'm going to find mine and faraday that shit this weekend.
Let me guess: you are afraid of the surveillance, use only cash for payment, pager for connectivity and have no permanent address?
It's safer and convenient to have a connected car. I like it that way. When you can open your car app and check the location where it is parked, amount of gas or request ventilation/ac.
That is a blatant reductio ad absurdum. People should have the right to reasonable privacy without being homeless and off grid.
I am happy to share my address and phone number etc. but not to be tracked wherever I go or have everything I do on my computer or within my home monitored. These are different things.
The industry keeps chasing "connected experiences" and ad monetization while ignoring what most drivers actually want: responsiveness, simplicity, and reliability
to be honest, we are now at the stage where everything that CAN get an internet connection, WILL get one eventually. Be it your god damn dryer or fridge or lawnmower...
Just bought a fridge. It was very difficult to find one with normal doors, a freeze door (not drawer), no stupid water or ice-making gizmos, and no wifi. There was literally one single choice in the dimensions I needed, unless I spent 6x as much on a European import.
You don't have to connect your fridge to wifi. Sometimes you can ignore a feature. Adding wifi to a modern fridge is really cheap (a micro controller is better than a mechanical thermostat or so I'm told - often wifi is included in a micro controller even if you don't use it)
It requires at least a basic cellular module for eCall in Europe since 2018, so car manufacturers use the already present hardware to provide more services. Maps and updates (live traffic view), internet hotspots for passengers (IIRC, Tesla does that one), entertainment that doesn't rely on a phone, firmware updates, feedback of driving data to insurances (yes, some insurances offer discounts in exchange for proving you "drive safely"), position data for leased/financed cars in case they need to be repo'd, synchronizing stuff like seat and mirror position across a fleet, remote pre-heating, "put packages in my trunk" access for parcel deliveries to thwart porch pirates, uploading data from real-world traffic situations to train AIs (again, Tesla does that one)...
There's quite the laundry list of nifty to nasty things that can be done with a connected car.
Let me ask it a different way. when the cell carriers turn off the xG towers and those features fail to work will you spend your own money to get the replacement controller or just do without
> giving me the current fuel and battery levels in the app
When is this actually useful? In the ~12 years I’ve been driving, I’ve never needed to know the fuel level of a car when I’m not in it. I guess maybe if I’m planning a road trip and need to know if I’m going to have to stop for gas before I leave? But I’ll figure that out when I get in to leave and I’m probably not leaving with <10 minutes of margin.
> locating my car
Again, never once have I not known where my car was. I think my phone keeps track of where I park too already? But I’ve never needed that feature. I guess if it’s stolen and the thieves don’t know how to disable this, it could potentially be useful for insurance/police.
> telling me about open windows, giving me the option to close them remotely
This could be useful. I’ve never left windows open by accident before, but I have left them open on purpose - if there were an automatic notification when this happens, I’d probably just eventually turn it off to reduce the irritation from false positives, and then not be notified if I ever left them open by accident.
> remote door un/locking
I had a Lincoln that had this feature, while I was working as a reverse engineer/pentester. Took me ~45 minutes to be able to send an unlock request to the car, unauthenticated, and have it open the doors, over the internet. Pretty sure that’s never been fixed (at least, it hadn’t been when I got rid of the car - model year 2016, which was identical to the 2013s, and I got rid of it in 2022). Needless to say, not a fan of that kind of “feature.”
I could see charging ETA being useful if multiple people are using the same car and for whatever reason can’t communicate that sort of thing with each other, and don’t have a feel for how long charging takes. (I’ve never owned an EV, but I imagine that you plug it in when you get home, and then it’s ready for you in the morning, so I don’t really know what the use case for knowing the ETA is in that case. Maybe if you’ve been driving around all day and need to make a long drive in the evening? I still assume you’d know how long it’ll take to charge when you plug it in though. And if you're at a fast charger, don’t they have a screen that gives you the ETA when you plug it in? I’ve only used one before, but it did that, and it was accurate to within 30 seconds, so I’m not too sure how useful it would be to have the ETA on your phone in that case either.)
Re charging times, there are several aspects here:
* charging stations have different powers
* charging time depends non-linearly on the remaining change
* ... and it's also temperature dependent (though only a little, with my plug-in hybrid)
I cannot plug in at home, but there's a public charging station around the corner, 3 minutes walk.
So I arrive there, plugin in, and the car gives me an estimation when charge might be finished. The initial estimate is sometimes off by up to 30 minutes (usually less). Sometimes I also forget the estimate, because I'm too busy with other things.
Getting a notification when the charge has finished, and an updated ETA on demand, is a notable QoL improvement.
It feels a bit similar to bluetooth headphones: I never complained about the cable before I switched to bluetooth. But now, I'd find it annoying to go back to cables.
In some areas you pay different rates to charge your EV at different times of the day. If that is the case for you then you will want to pay attention to and control charging. Normally you will want to do a get your car charging at the cheap rate, but sometimes you are willing to pay triple costs because you need to leave soon and there isn't enough remaining range. (your car matters - my PHEV only gets 30 miles from a full charge, so I'm always low when I get home, while many EVs get several hundred miles and so really only run low when on a road trip)
Eventually I expect cars, chargers, calendars, and the electric company will somehow integrate so that you can plug in and the system figures out when to charge your car. that is a complex project though with privacy concerns that are hard to address.
The Lincoln thing reminds me, there was an article posted on HN about how a researcher found a vuln allowing him to track and unlock any LTE-equipped Subaru.
There are several reasons to have internet connectivity in a car. For example, you might want to start your car remotely (e.g. winter time and you want to pre-heat it) or you'd like your onboard navigation maps to update automatically, or you'd like the latest traffic reports (if available in your region), yadda yadda.
While there are a lot of people that love Android Auto and Apple CarPlay, there still remains a sizeable group of people that want to have a navigation solution without using their phone, or to be able to enjoy their car without having a smartphone at all.
Software defined vehicle? Never heard of this term. More marketing buzzword BS.
Yes, Tesla has one of the best user interfaces in a car, and has set the bar high. But just because they have OTA updates it's now called a "Software Defined Vehicle"?
From first principles I think the concept can make sense. From car-specific function-specific ECUs, to platform-shared (but still function-specific) ECUs, then to Zonal architecture and domain controllers. The goals: consolidate and generalize HW across the lineup moving model-specific bits to FW/SW/Config (amortizes the development cost and simplifies certification), and also simplify wiring (saves you precious copper wires which are costly, messy, and heavy) because you can pretty much just plug every miscellaneous sensor or actuator to its nearest "anchor point" without worrying (too much) about arbitrary ECU limitations.
This might sound like purely implementation detail, but having the (non-safety-critical) "business logic" of a car as software gives the manufacturer flexibility to late-bind behavior as new use cases / demands inevitably get discovered.
Something can simultaneously be a good idea, buzzword'd by marketing, and/or deviate from the original intentions.
General Motors was in the lead then they just quit. It was stunning to see all of their incredible self driving Cruise cars vanish and then overnight see them all replaced by Waymos. It was like watching the downfall of Xerox PARC.
I've long wondered why no car manufacturer has gone for an open source model. Certain things should absolutely be locked down (for example, the airbags and other critical safety features) but there is absolutely no reason the HVAC and Infotainment system need to be closed source. Open it up and let hackers go crazy, then just "borrow" the best options out there for next year's model and everyone wins!
I have been wondering the same, but slightly differently.
Tier 1 suppliers have enough resources in both know-how and manpower that I have been wondering if they could do a platform car. Provide a basic frame that passes crash, provide a basic engine that passes emissions, provide basic safety, etcetera.
Then invite other parties to upgrade components. Package lots of air between components to simplify compatibility.
I suppose the only way to get this going in the real world is a big military contract, but I am wondering if it wouldn't be smart play for everyone involved. It would be deadly for a bunch of traditional automakers, but they can't do anything preventing it.
Automotive-grade Linux is actually a pretty big thing but cars being put on the roads still need to pass through approvals. It's not "hackers" doing anything they feel like.
There could be a sort of "ARM" or "Android" but for cars.
Come up with few general hardware modules, enough to replace the head unit, body controllers, ECU, climate control, and ideally driving automation, and software to run them. Everything minus safety modules like the airbag controllers, and then license them under Fair/non-discriminatory terms.
Then, a variety of automakers get access to core functionality and cheaper hardware to run it. That means that the cars themselves can have higher quality software, cheaper hardware (from cutting out companies like Bosch that charge exorbitantly for things like a windshield wiper controller), and thus deliver more value to customers.
Apparently there are folks working on converting ICE vehicles to EV, using open source and reverse engineering to integrate discarded proprietary EV components (battery, motor, inverter etc).
Because there are regulations for the infotainment system as well. For example, you can't watch videos or read SMS/messages. Not to mention that the infotainment system likely has access to the CAN bus, through which you _could_ impact other safety systems.
From the article: "Evidence of that dichotomy is not hard to find. As automakers have introduced vehicles with more advanced computing and electrical architectures, they have also struggled to deliver bug-free software on time."
This was something that really hit me when the Internet allowed game developers to ship a game that wasn't done. You got the game, and the first thing you did was download a "patch" that was at least as big as the CD the game came on (several hundred MB). I've got "released" Windows98 games on CD that are essentially unplayable because what was shipped on the CD was unplayable and without the update server on the network sending out those critical fixes, its never gonna work. For game archivists that means finding a fully patched install and then preserving that.
This is a shitty experience that serves manufacturers but not their customers. I don't expect it to get better any time soon but I wish it would.
It's not just about the software but about the hardware architecture of the car. Legacy manufacturers are coming from a situation where they are integrating hardware and software from a lot of different suppliers. This makes upgrading the car a very tedious process and slows down the process of getting suppliers to fix issues and provide new firmware in a timely fashion. It's worse for them because they often want to do ICE and EV variants of the same car. Which means sticking with the same supply chains and associated issues.
Vertically integrated companies do this very differently. Tesla pioneered this. The Chinese copied this and at this point you also have companies like Rivian and a few of the legacy manufacturers that are doing the same. Effectively they in house all the software and e.g. Rivian runs the software on a handful of hardware subsystems instead of having hundreds of chips with their own firmware for things like the wind screen wipers, the software that controls the windows, the AC, the keyfob, AI driving features, and so on.
I mention Rivian here because they just did a deal with VW to start doing the same for them.
The issues here are not just technical but cultural. I used to work in Nokia when it was in the (slow) process of figuring out that they were a software company rather than a hardware company. Then Apple and Google came along and they were slow to adapt their internal processes and management. Apple makes firmware that goes on their phone. They provide OTA updates. There's only one supported version of that firmware: the current & latest one. It's the same for all phones they still support with updates. Nokia did the opposite. They forked their software for each product variant (dozens per year). And they did not do OTA upgrades. So most of their phones weren't updated at all (by users), and would typically ship with bugs that had already been fixed on other branches of the software. And it would ship on the schedule of the manufacturing process, regardless of the state of the software. With all the obvious consequences. Nokia got a well deserved reputation of shipping half baked software.
By the time MS bought them out, they had learned and improved a lot but Apple and Google were running circles around them by then and it did not matter anymore.
You see the same with car manufacturers currently. It's all about the buttons and the bling. They have a gazillion of upsells, features, special trims, and what not. And it all adds up to a whole lot of nothing if the software experience isn't great. That's why VW is paying billions to Rivian to fix that for them.
Their cars are too expensive, have too many chips and wires, and their software just isn't good enough. And they don't have ten years to figure this out for themselves. That's what Rivian is supposedly fixing for them.
Legacy corporations have a very hard time incorporating fundamental technology shifts (moving from ICE engine/drivetrain dominant designs to software dominance).
They walk into the future looking backward, unable to identify/vet the team skills needed going forward, leading to silly hacks like: 1) hire from "big s/w companies", 2) pay high salaries to poorly vetted people, 3) adopt all the new fashionable buzzwords like "software defined vehicle", 4) force new teams every inch of the way to justify design choices to mediocre legacy management.
The only formula I know that works is "hire good people and listen to them".
From experience, the only way legacy companies can do this is acquire and/or seriously partner with companies that have established a track record in what you need (even if it is only a couple of years, as long as they are _delivering product_).
As software effectiveness/innovation speed/productivity continue to increasingly crush legacy industries, it is extraordinarily frustrating to see how hard it is to make (seemingly simple!) changes.
Cars now have computers, cellular internet connections, cameras, microphones, privacy policies... I can barely find the words to describe just how frightening the status quo is.
Indeed. Reading the comments here makes me a bit more grateful for my early 2010s vehicle. I added a Bluetooth module so I can play music wirelessly. My phone magnetically connects to an air vent and starts charging. I open Maps and tell it where I want to go. Done. :)
I can confirm that Volkswagen is borderline incompetent when it comes to software - a few months back, my 2020 Audi A4 (and those of tens to hundreds of others) all started having the same issue, where the infotainment will randomly reboot every 5-30 minutes (taking out nav, the backup camera, and the parking sensors with it, and requiring a PIN to get back into the system).
Despite the problem having the hallmarks of a backend issue (many cars with the same software running into the same issue on the same week), corporate is still insisting that it's a hardware issue and trying to sell us on $5k hardware replacements. I love the car for its build quality, but almost kind of wish I'd gotten a Tesla given how bad VW is at software.
Pretty cool seeing how all those little gripes with car tech stack up, kinda makes me question if adding more software actually makes things better or just adds more mess. you ever feel like simpler is actually safer when it comes to stuff like this?
Just talk to Canoncal, or IBM, make a NixOS config, or just do something. How hard can it be? My father’s 5 yo Volkswagen van has an 80’s looking UI, the touchscreen is already failing. Going from the normal UI to CarPlay is just jarring, any 2024 Linux distro looks, feels and acts more modern. What are they doing over there??
I could probably whip him up something nicer if only there was just a Nuc or something in there somewhere.
While I don't dismiss your general point, I will say that anyone who says "how hard can it be" really needs to consider they are falling victim to the Dunning-Kruger Effect. In my experience, that phrase is (typically) a red flag for the latter.
This strikes me very much as one of the things where the answer is probably very simple but also very difficult.
I would also guess (completely un-informedly) that because the simple (and probably correct) answer is very difficult, a lot of companies are trying to avoid it by doing things that are more complicated but also easier. And because they are more complicated, it is not immediately obvious why they won't work....but they won't. Which is resulting in the repeated failures.
I'll let you know why I finally get my own van, add DC powered Nuc-like with a touchscreen and just run Gnome on it. Or, to make it more fair, I'll put LineageOS on it. Very touch friendly.
It isn't just the UI. You need to (re-)read the article. They have a bunch of interconnected systems that need to talk to each other, the existing methods are fit for purpose.
These systems also have to work correctly (100% of the time) in a range of conditions and it needs to not drain the battery while the car isn't being used. They also need to start up quick on relatively low end hardware. The car creates a very hostile environment generally for electronics. There is lots of dirt, muck etc that will literally work its way in everywhere. There is also a bunch of regulations that have to work almost internationally.
Car companies are not software companies. If you are a software developer not in a software company, things are much more difficult as the organisation just isn't geared to deal with software development generally. Combine this with it being a massively complicated product (modern vehicles are complicated) you are setting yourself up for failure.
IMHO, a computer car and even internet-connected car is fine. However, I want a computer car that I actually own. If it's my car that I paid for, I should have full access to the software that runs it. If not, then I don't own the car, I'm just renting it.
Yes, Titan was cancelled in 2024 after eight years of development. There's a good breakdown on The Information about why, it mostly boils down to software challenges (especially ADAS), leadership turnover and a fair amount of internal skepticism that it was even a worthwhile project.
Apple had a secret test track in Arizona, with buildings made from shipping containers. You can see it on Google Maps under "Chrysler Oval Track".
846 comments
[ 2.6 ms ] story [ 370 ms ] threadTo be fair, im still not sold that this is an advancement except maybe in simplifying the number of components. I'd prefer the car to work without "updates" and DLC. Why does my car need a firewall??
[1] https://wikipedia.org/wiki/Firewall_(engine)
For what it's worth, I work in this industry and the general rule of thumb is that every increase in validation from QM (standard quality) up to the various levels of safety critical code has up to 10x the cost per line of code of the previous level.
Exactly that was done for decades.
Heck, manufacturers were issuing service bulletins to fix the fuel maps in their cars in the 1980s.
Now that all vehicles have entertainment systems connected to the internet, I guess it is tempting to use that to reprogram ECUs, I haven't been working in this area recently though.
The first use case of connecting entertainment systems to a vehicle bus that I can remember was to read some engine settings and turn up the volume on the radio at higher speeds.
Is anyone actually begging for this though? And why do you need a full bus? This feels like a luxury car problem that could be solved over I2C or something.
I’m reading this whole SDV thing, and outside of using less ECUs, it seems like an overengineered solution to what was hardly a problem. If we can update ECUs already with OBD-II, step 1 is just making a virtualized OBD-II port that the infotainment system can talk to. Everything else can then stay unchanged until later.
A "virtualized OBD-II" is really just a UDS server if I understand what you're trying to convey. UDS is a dumpster fire of a protocol that should be expunged from existence, but my personal feelings aside can be run anywhere you want. That exists. I'm not aware of many systems that directly connect the infotainment processors directly to critical CAN buses. Usually there's an intermediary component to isolate them.
Yes, but code that doesn't get written does not have bugs. And I don't want to control the rear window defroster, wipers, climate control, fog lights or whatever, on a touch screen menu buried 7 levels deep while going 130 km/h. It's bad enough that coffee makers, light bulbs and tooth brushes now have updatable firmware.
Why? If the rest of the car can function within design specifications for years, why can't the firmware?
I'm fine with updates to add compatibility with new protocols and such, but to me a bug implies there's a standing problem with the current system that's not due to some sort of wear/changing standard/component damage etc. While one can point to examples of cars with defective mechanical designs, I don't think anyone considers it impossible to create designs without such defects (where defects are defined wrt. specifications), why is this the view in software engineering?
But, do you have an example of a software project anywhere that's bug-free? I'd include the space shuttle code, but even that famously high quality development process produced a (low) number of bugs.
The reason for this is a physical limitation: the cars weren’t shipped with wideband O2 sensors, so there’s no way to measure the AFR when wide-open (since it’s targeting a significantly richer mixture, and narrowband O2 sensors can only signal whether a the combustion is stoichiometric, or rich or lean relative to stoichiometric, with no further info). The implantation is probably not a bug but rather a compromise; in an ideal world, the “most recent” BLM will hopefully be from an “almost wide open” part of the map, and the general rich/lean characteristics will be close enough. And, the fuel table in the factory tune is quite safely rich when wide-open, so even with a leaking injector causing the idle BLMs to be way off, the fuel being pulled when wide open will still be completely safe.
Aside from that, 128k of bug-free code.
What was wrong with ECU and ABS etc software prior to the OTA era that we're now apparently entering?
I've had plenty of cars--too many--and outside of a few warranty repairs involving re-flashing ECU/ABS(maybe), this was a very rare occurrence.
(Not counting deliberate tunes or re-flashes for modification purposes)
One, it's expensive. If your update takes half an hour to apply, under the old model someone's being paid half an hour to apply it. Either the manufacturer cuts the billable hours to the dealer and the dealer loses, or the manufacturer is paying that half hour out of increased prices to the consumer. With an OTA system there's usually no cost to anyone besides network traffic. This amounts to billions of dollars in savings for manufacturers.
Second, owners hate 1) paying for updates and 2) getting notifications about it in the mail. It generates bad press and bad experiences for the manufacturer.
Three, it makes the production line more efficient.
Four, the old systems sucked to maintain and for techs to use. They were also insecure and retrofitting security is impossible in a standards compliant way. The internet people have done a much better job with their standards.
Five, most owners are not like you and I. It's a feature for them that their car gets improvements and fixes automatically.
Six, you can be pretty certain what the rollout distribution is. Regulators don't like it when owners are driving around with years old recalls active because they forgot to schedule a dealer appointment. Manufacturers don't like keeping the inventory around.
Seven, "networked services" can piggyback on the same infrastructure and provide additional revenue streams. Certain corporate types think of this as one of the main benefits. Remember how manufacturers used to sell annual maps updates that no one bought? Some consumers also enjoy these sorts of networked services, which frankly I find a bit baffling.
Hmm, I disagree. Bug-free systems are expensive and hard, and get more expensive and harder as complexity increases, but you can absolutely produce a car that never needs updates. The vast majority of computer-controlled cars from the 80s to the early 2010s never needed updates, and the ones that did were performed at dealers (and were usually for non-critical things, because the critical things were simple).
GM had a good run in from the mid-90s to the mid-00s producing bug-free cars, even with some complexity. I don’t know of any software issues on any cars with LT1 or 3800 engines, nor with any of the tech in the Northstar Cadillacs. Displacement-on-demand could be considered a buggy implementation, but it was working as designed, and never got patched out, so I don’t think it counts.
That’s of course ignoring the decades of cars that had no computers at all. No software bugs being patched out with OTA updates in a carburetter (you have other problems obviously though, namely terrible fuel economy and emissions, and generally lower reliability).
If you make it a hard requirement for a car to be bug-free (maybe outlaw OTA updates and force physical recalls on any software problem?) I can guarantee manufacturers can make a bug-free car. It’ll just be way less complex and have way fewer flashy features, and will either cost more or have lower margins. It’s been done in the past, it can be done again.
There is a sweet spot for the level of computerization in cars. We had it somewhere around the year 2000, then waaaaay overshot, and haven’t corrected back.
Alot of embedded stuff is outsourced and doesn't want to waste the computing power for stuff like stack canaries. I recall the following from making a tool for a dlink? router?
//Reads a file name foo ReadFilePath() { // Get file name // TICKET 21321: Fixed crash by increasing buffer size char FilePath[100]; ReadFileName(&FilePath); }
It sticks out to me, since the crash was clearly from a buffer overflow, and they had this documented in the source code that increasing the buffer size fixes it. What they didn't realize was that the bug would still happen and you could get a buffer overflow from this and do whatever you wanted. This is the level of programmer you're dealing with who's writing embedded software in an overseas sweatshop. And the talent isn't even there domestically since they're severely underpaid compared to someone writing simple javascript.
The pipelines to create more such people are sorely lacking though
But if you take a couple C/assembly/systems electives, look for internships at hardware companies, build a couple of toy projects on the side, and graduate with even a modicum of embedded experience, there will be companies that will hire you, pretty much guaranteed. You won’t be making 250k out of the gate, but you should still be making a more-than-livable salary (and frequently in a lower cost of living area than, say, the Bay), and if you pick companies correctly, you can be working with and learning from some truly genius engineers.
The pipeline’s there, it’s just not attractive (read: $$$$$$$) enough to pull in most people in the industry.
But yea, a single class probably isn't sufficient and also I image a lot of embedded companies have a preference to hire someone already familiar with the chip they are targeting and the toolchain for the stack. I also see a lot of asking for experience with RTOS, which in my class, we didn't use an RTOS.
I did some initial requirements work on a system to monitor continuous-web papermaking machinery; the line had to be stopped, physically and completely, within 100ms if anything went wrong, because an uncontained web of paper can literally cut people in half. They wanted, in order to be able to hire, to use one of the embedded flavors of a well-known consumer-grade OS, and I had to prove to them that there was no way to make any of them safe, at any cost. And they knew their hardware, because they had built it themselves.
The absolute last resort is a watchdog timer that hits the reset button if N milliseconds go by without the software telling it it's okay. This is what you have to implement if you are dealing with buggy and undocumented hardware -- as, all too often, you are. Sometimes you can get some doco for $ and an NDA, but then in order to get the real doco it is much more $$$ and a much tighter NDA, and the existence of that option is not even divulged until after things have already gone very far south.
If it were only a matter of reading the top-level doco for this or that chip, there would be no issue.
If I were selling hardware I’d want it to be as open and well documented as possible. So that more people buy it and so that I get credit for all the great stuff people make with my products.
1) The more you open up your design and its behaviour, the more your competitors can learn about your product and how to possibly improve their own. Even stuff as basic as what specific features/capabilities a specific SKU at a specific price point has can be useful information.
2) The behaviour may be sufficiently undefined as to make documenting it impractical (or a bad look). Specs may also be padded (“up to 14 bits of SNR” may mean you’re getting 8 most of the time unless you’ve got a golden sample, and you’re not getting the distribution without paying big bucks and signing a big NDA). This ties in with 1) - if your competitors know your exact yields, they might be able to advertise being better/more reliable more truthfully, or even cheap out on their manufacturing a bit to drop their own yields down to match or just slightly beat yours.
3) The behaviour might be unknown. There’s obviously a crazy amount of validation testing that goes into high-end chips, but even the best test plan can miss things. This is especially true when you’re talking about high-speed stuff and anything involving power delivery/voltage fluctuations, or async/pipeline executions, or a million other things that can go wrong. Again ties into 1) - if your competitor knows that your chip might deadlock the radio with an obscure pattern of inputs and control signals, that could give them insight into how you’ve laid out your silicon and might give them optimization ideas.
4) If all the available info is given out freely, then potential customers can easily compare manufacturers and pick the best one. The manufacturers don’t want this, unless they’re the best, for obvious reasons. And, because everything’s locked down so tightly, no one knows if they’re the best until the chips are on the market and the volume contracts are already signed. And those contracts are hard to break, since the specs agreed upon are pretty vague due to 1-3.
5) The manufacturer knows their chips suck, but needs them moved anyways. This is rarely the case from most non-discount manufacturers, but it can happen. In this case, you don’t want to give away anything you don’t have to, because most info you give out is going to drive customers away to a better option. Good example in the consumer space is Intel refusing to publish acceptable voltage specs for their 12-14th gen Core chips, which resulted in motherboard manufacturers overvolting and killing high-end CPUs to try to meet the frequency specs Intel was advertising. If Intel was truthful in their voltage and frequency specs, there’d be a minuscule percentage of chips that could actually hit the advertised frequency at safe voltages, and 99% would have worse performance than expected, which would almost definitely result in lower sales.
6) The behaviour may be highly dependent on external factors. Basic example, a chip with external DRAM might have its execution pipeline stalled more or less frequently based of DRAM spec, or a wobbly voltage regulator might be known to cause lockups when certain executions are happening. Are you going to tell your customer those problems, or just say “we recommend high-speed DRAM and high-quality VRMs?” Especially if the other guy just says “we recommend high-speed DRAM and high-quality VRMs?”
The world would likely be a better place without such logic, but the incentive is there. Until someone comes and breaks the paradigm, I don’t see things changing.
This is an industry that is about as far from the light of science & enlightenment as it is possible to get, ensnared as deeply in the entangling anti-human anti-science Intellectual Property qualgmire-hell as can be got. Oh sure plenty of science goes it! It's fantastically interesting & technical! But aside from some Application Notes write-ups trying desperately to help move the practice along, move it out of jank, knowledge goes in, but it doesn't ever come out! There's such a lack of peershios with which to practice science, to report your findings to, to replicate works on.
The software world talks about its patterns and practices. The biggest industries on the planet are building software like wild AND are mad into open source. But... computer engineering is the shadowland, where no talk nor victories that happen there are allowed to be shared, where nothing escapes confinement. What a fucking plagued awful land of people unable to ever do the right thing, unable to bring their work out of the dark & into real civilization.
What does writing ABS module software look like? I'd actually love to know--it's not an area where you can "vibe code" your way to a 'working' product.
The process is so far removed from typical web and business slop that it's an entirely different world of its own.
I'm waiting for a recall fix for the underpowered Sync 2.5 system to correct a backup camera problem. I'm not looking forward to worsening of all the current bugs with USB audio file playback that cause the UI to hang or not show a fully rendered display.
can someone tell me if there are any course that taught this??
Maybe things just really suck for embedded in the states? But since my last year of university I’ve been inundated with recruiters for embedded positions, and I’ve never had a problem finding work. ~75th percentile in salary alone for software engineers in my area, ~55th-60th for Canada. I make more than every JS developer I know who graduated with me, except for the ones who moved to Seattle, Vancouver, or the Bay.
Now sure, if you’re looking for 500k+ jobs, embedded isn’t the area to be in, unfortunately. But I prefer low-stress, fun-environment embedded jobs, and don’t mind trading off salary for that. Different strokes.
At this point, when I wanted to get back into hardware, it made more financial sense to outfit my home office with all the measuring instruments, debuggers, tools and other equipment necessary for embedded work and do it as a hobby. If I had the space, I could even get full-size CNC machines and still come out ahead cash wise. It’s insane.
It’s no wonder they can’t find experienced embedded devs, when it makes no financial sense to stick with it over a decade.
Part of the problem might be poaching high title people from embedded tech companies while not doing anything for developer compensation.
I’d probably add that the pay scale for software vs. electrical/mechanical people probably wasn’t notably different in the 90s or so. And California rates didn’t compensate for CoL in general. Very different.
New car is basically a computer on a simple chassis with an equally simple drive train. Software and battery tech is everything.
OTOH, ABS and ESP systems can achieve similar or even better results with less complexity because motor torque control is inherently low-latency, which can also complement brake deployment (hydraulics is not as well behaved as e-motor).
You do get rid of emissions control and tiny little sensors / flap actuators sprinkled all around the engine bay, so yeah, probably overall still a simplification win, but I doubt you can get very far without "massive amounts of [Mechatronics] engineering".
Both Google and Apple have car software, and who knows if Apple actually developed a full stack of the way Tesla did. But anyone can download and play with android automotive. It's unclear what getting one of the android automotive developers would do for you.
Whoever convinced the people writing requirements documents for car user interfaces that they needed to use Unreal Engine to show you what your own car looks like and spin it around in. 3-D deserves some kind of salesmanship Nobel prize. That is the most pervasive useless thing I've seen in a long time.
And so did traditional manufacturers, they just had the benefit of being able to phase it in if they so chose. Or they could have done a hard cutover, either way, the failure is on them for ignoring the benefits of the Software Defined Vehicle discussed in the article.
> It's unclear what getting one of the android automotive developers would do for you.
Do they do vehicle control systems or just infotainment?
> they needed to use Unreal Engine to show you what your own car looks like and spin it around in. 3-D deserves some kind of salesmanship Nobel prize.
I mean that's exactly the kind of thing that makes Tesla fanboys rave endlessly about their car. It just needs to be decoupled from the actual software system, like any UI.
All that stuff adds up. As Volkswagen found out.
Green Hills supports running android in a VM so you can do all of the safety critical things like traction control, and ABS in a secure environment.
Now I read that Stellantis is behind on the software game and I wonder if there is a relation. Seriously, I'm all for cost-effective cars but reading the article I do not get the feeling that so-called SDV are in the interest of me, the consumer.
Also, since I've worked on military systems a lot, I suppose a military grade firewall is just iptables for which someone has written a shitty gui (that might as well just be a webshell) and packaged it into a green rugged box.
Consider this. Almost every car on the road today has an unsecured bus going back to like the 1980s. However you need to actually access the car to do something malicious so the threat vector is zero; since if you have access to the car you can also just cut brakes or put in a pipe bomb.
The only reason why this paradigm changes in the EV era is because the insistence on having EVs phone home. Now you can concievably hack all EVs of this model at once and that is now realistic and even attractive to do. But again not a necessity for running a car. Just something that modern software focused companies want to see that leads to a host of expensive security issues that didn’t exist before. The car could be airgapped with the dealer network used to flash software updates like they do with most other cars before EV era.
Sure someone in that situation could also "just cut brakes or put in a pipe bomb" but car theft is a lot more common than assassination, at least where I live.
See [1] from 2023, where popping the headlight gives access to the bus. Lack of internal security then gives a way to steal the car.
The threat just isn't the same as the one you are modeling.
Security will come eventually, if only to prevent bad publicity.
[1]: https://arstechnica.com/information-technology/2023/04/crook...
ETA: Just as the sibling says...
I would rather have OTA updates than enable parasitic middlemen to siphon money out of me
"military grade" is often used as a marketing term used for things that pretend to be built to be extra strong.
In this case it is a stupid term to use to describe a firewall cause a firewall either works or it does not.
Data streams are converted into a sequence of objects that are required to have and satisfy certain formally verifiable properties as a pre-condition of forwarding. Any data or objects that cannot satisfy formal analysis requirements are dropped. Forwarding policies are only applied to objects that meet the prerequisite of being rigorously analyzable.
This behavior is bidirectional. It applies equally to data egress to mitigate internal threats and accidental data leakage. The internal mechanics can be pretty complicated and they necessarily operate on a store-and-forward basis. The data objects may be “laundered” by the firewall, what you send may not be exactly what the other side receives.
To make this work, the wire protocol, data representation, etc must be designed specifically to allow this kind of rigorous analysis and work well within these constraints. It usually won’t work on a random web stream and the data representation often sacrifices efficiency of storage for efficiency of verification and analysis at runtime.
In reality, virtually no one uses this type of tech outside of defense and intelligence because it won’t let almost any of the standard web stack slop through.
[0] https://en.wikipedia.org/wiki/Unidirectional_network
I don’t know what constitutes a “military grade firewall” but presumably something that stops that. Or at least tries to.
No clue about firewalls though.
Some unexpected Kierkegaard in there (I only recently learned Dune was referencing it).
[1] https://en.wikipedia.org/wiki/Lada_Niva
Part of me thinks the reason they are doing an integrated system is a combination of economics and convenience for 3 letter agencies to remotely assassinate ppl.
Basically sitting inside a Windows that can kill you.
They all lost their minds putting stakes on software makers. I intentionally avoid the word engineering, engineering is far far away what is built up by the software making industry that is now tasked with being the babckbone of vechicles you put your and your family's life into. The cultures are incompatible.
(disregard mission critical software, their engineers are not proud members of the 'do not finalize, fix it later' bunch, not at all, they are nowhere here)
Software for steering or braking systems is of high quality. It's not the same team that does the infotainment.
It states that consumer reports, (a for profit company providing independent reviews, and not a regulatory body) said the Model 3 stopping distance was not good. Allegedly due to a “bad ABS calibration”. Tesla released an OTA SW update.
Why wasn’t the bad calibration and degraded performance caught by regulators testing automobile safety standards?
The article also posits that this ability to make OTA updates expands the (IMO very very bad) SWE perspective that “it’s OK to ship unfinished and buggy products” into safety critical systems.
It sounds like their ABS system wasn't designed as carefully as conventional systems if there was such poor braking performance. Reading around, it might have been related to the emergency brake assist functionality not being calibrated properly.
Another consequence is that ISO-26262 and most other standards are completely, 100% norm-based in the US. They're used because the industry expects them, not because there's a legal requirement. You can deviate all you want and the only consequence is that regulators might take a closer look at your paperwork in the event of issues because they look unusual.
I disable the "land assist" every time (which often tries to steer me into wildlife or other cars and was clearly not built for use on a single track country roads with hedges and random verges), but this was the first time in 3 years that the "front assist" caused problems.
If that's "high quality", I dread to think what low quality would be.
I'm one of them. Yet I still haven't had a situation where "lane assist" or "front assist" has actually been a good thing.
I am now of the opinion that a car should never under any circumstance drive for you. If a car has cruise control it should cruise control you into a wall. That I can at least anticipate.
The decision to do an emergency break is the same problem fully self-driving cars need. You need to interpret sensory input and have a model of the environment.
Ironically some genius made these systems mandatory despite them being a safety concern. Granted, they tend to work if someone really falls asleep behind the wheel.
On one of our cars, it is fine. On the other, it’s so bad they should have to buyback the last N years of vehicles. I hear the same high-variance story from friends.
The standards for this stuff are completely inadequate.
That is a piece of paper.
> Software for steering or braking systems is of high quality.
There's literally no way for me to know that before I trust my life with it.
I do not. A more charitable way to phrase that is "We are all expected to." And yes, well spotted, this problem extends well beyond vehicles. Or are you suggesting that this is somehow indicative that there are no problems? How would we all know if there _was_ an error in a device?
> those that conform to “piece of paper” standards, such as ISO 15708
That standard deals with non destructive testing and has no material that is related to the practice of medicine or the use of medical imaging scanners. It's not even the right piece of paper.
Is this then logic that gets airlines to buy from The Boeing "Are door plugs supposed to stay in?" Company?
In the fully autonomous future the car I want to own and drive will still be my 6MT 911! :-)
If I want to be driven, I’ll just book a waymo.
So move to one of the 2 or 3 cities in the US that have Waymo?
We aren’t there yet.
Also, you can just buy older cars, that works too.
BTW, I thought about buying a Lada Niva, because I love the looks, but I heard it is not that reliable as you would assume, and they are pretty pricey for a car that is basically the same for forty years…
Anywhere I can read more about this? Sounds terrible.
As I understand it, yes the system worked as designed, but the design still managed to kill several hundred people.
I'm not qualified to evaluate the design of the system itself. Was it inherently flawed or would everything have been fine if the optional backup sensor had been mandatory, making this another example of corporate greed causing tragedy?
Either way, I don't think blaming the pilots is fair.
Definitely not trying to be Boeing friendly fwiw.
> a flight control feature the pilots didn't know about
phrasing that made me take issue. To me, it implies that if just the pilots had known about MCAS in detail, the crashes wouldn't have happened, implicitly placing the blame on them. While the reason for both the faulty system and the lack of knowledge can be attributed to Boeing trying to save money, which I think is a more fair target for blame.
So, is it really sane to put similar features in cars, where you get your driving licence at 16/18, and then that's it?
This also goes for the huge screens on the console. A pilot has been trained for each commercial aircraft model they fly to navigate their way around the numerous controls. But putting a tablet in front of an untrained driver? It sells well because it makes you feel as a pilot. But at the same time, it is a huge distraction and there is zero training to cope with it.
Is this how we get the Butlerian Jihad? Because part of me sure does want to learn how to identify cars built like this and learn ways to disable them when I see them parked somewhere around town, before one of them fails to recognize me on my bicycle as something that should be avoided.
https://www.slate.auto/en
Give me a car that is perfectly 100% autonomous, or give me a car with three gauges and basic controls only. Everything else is an uncanny valley: all the downsides of complex tech without being useful enough to justify it.
Until then I like my Nissan Leaf: physical controls, phone just docks with infotainment screen, and reliable.
You can an intuition pretty quickly for what it does and what it doesn't, and in certain situations it really takes a lot of attention off your plate (stop-and-go traffic, and long distances on the highway).
Edit: maybe my information was old - some sources say it costs nothing
USB-C is so powerful, it can do everything Bluetooth does while charging, but for some reason that's just not an option in a lot of cars? Make it make sense.
This is with USB, too.
I want the car to start and CarPlay to be operational; we have no time to be wasting on whatever formalities software wants to have.
Maybe someday wireless CarPlay could start syncing with the system before you even get to the car, so it's already loaded when you sit down and start.
Also, during short stops, the screens go black but the connection is kept up, so when you re-start, there is no delay.
Whereas I really did take wired CarPlay into consideration when buying our minivan, there are only so many options that I may have had to compromise.
It has a phone holder where other trim levels would place the screen, and USB power around there.
Other than that, the car is mostly Bluetooth a speaker.
They actually have an app that allows you to tune the FM radio, otherwise I don't think you can listen to radio broadcasts.
https://play.google.com/store/apps/details?id=com.dacia.dngo
“These guys do it…”
“Oh no, not those guys!”
We could instead discuss the feature, but no.
I just mean I'd totally buy a much higher end car that is like this, I don't need a screen with all the nonsense on it.
Give me a car with no computer, but a phone stand and charger built in!
Oh oh, we could even use a standard like monitor stands.
Carplay is your phone, it's like having an external monitor
It does cost time/money to integrate, like any feature
I was annoyed enough that our used/new-to-us 2020 vehicle only supported wired that I bought a wired-to-wireless adapter and brought it with me on test drives to ensure that whatever I bought would work well in wireless mode [or else I was buying a different car].
I installed a wireless charger under one of the cubbies that was well sized to hold my phone on long drives. No need to faff around with cables.
Yes, for the main reason that I have a Starlink Mini on my roof rack.
My phone can connect to the vehicle via wifi, or it can connect to the internet over Starlink via wifi, but not both simultaneously. With wired CarPlay, that problem is solved.
You can't, it's required for eCall which is a mandatory feature in Europe.
Unfortunately, it's fraught with issues, especially for the very first eCall modules where the hardware supported only 3G (HSPA)... which is being phased out across Europe together with GPRS (1G)/EDGE (2G), leaving these cars without a working eCall system - and no upgraded hardware modules in many cases.
Unlikely to happen, but possible (not 100% safe, but good enough).
But remote controlling the car is something different.
That’s…terrible
Hopefully those same people know what ANPR is and how does it affect them.
And “modern” is going back over a decade. So most cars on the road.
Other than having some vested interest in denying people's right to choose between positive and negative aspects of progress, or having a Stockholm syndrome towards those who have such interest, I got no plausible explanations to this psychological phenomenon
You seem to be wasting your precious “armchair psychologist skills” here on hn.
And of course it's a tracker. It reports my location to a third party. There is no other definition for it. That it purportedly only does this during an "emergency" is not something I can verify nor trust.
Yes: it reports your location to a third party. When you have an accident. So you can get help.
And of course it can be verified. At the end of the day it’s a sim card in a passive mode. Maybe you don’t know how to verify it. There are millions of them driving around. I’m sure someone more qualified than you would have already reported if they were tracking in real-time. The system is nearly a decade old.
Storage is cheap. We know the NSA hoovers up everything it can and puts in a database. This is facile reasoning.
> So you can get help.
How many accidents resulted in fatalities because 911 couldn't find the victim? What's the percentage on that you think? Is there _no other way_ to get help other than this system?
> I’m sure someone more qualified than you would have already reported if
You're just full of confidence. Pray tell where it comes from. What makes you believe this is true?
I trust in people who look for this stuff as a daily job, or because „they’re here to get you”. Because you know a sim card in the open is really easy to verify. And there are thousands of them driving around but also easily approachable sitting targets in every possible setting.
But you double down. I find this discussion pointless.
> How many accidents resulted in fatalities because 911 couldn't find the victim?
You’re missing the point of the eCall system. Think out of the box. What if there is no witness? What if everyone in the car cannot get help
Everyone who could reach out for help, received it. Can you tell the same about those who weren’t able to contact first responders in time? Maybe they’re still missing! It’s called survivorship bias.
Sounds like you are trying hard to be obnoxious.
Can you explain why these protections are not sufficient for privacy?
> 112 eCall is not a black box. It does not record constantly the position of the vehicle, it records only a few data to determine the position and direction of the vehicle just before the crash and these data are only transmitted to emergency call centers if there is a serious crash.
> eCall cannot be used to monitor motorist's moves. The SIM-card used to transmit the eCall data is dormant, i.e. it is only activated in case the vehicle has a serious accident (e.g. the airbag is activated).
I know some carriers have strange quirks in their SIM provisioning systems. For instance, it may take more than a few minutes or require a specific type of coverage (like UMTS) for an otherwise "dormant" SIM card to activate from limited service on short notice.
I found an article about eCall Callback that confirms this is a known problem: https://eena.org/blog/resolving-the-ecall-callback-issue/
That statement is factually inconsistent. Either 112 eCall incorporates a time travel device or it must constantly record the position and direction of the vehicle and other data. In theory, that data is then deleted, but you have no way to verify that it is - and it would only require a trivial, unnoticeable software update to modify this.
Thankfully, we're safe. Car software is notoriously high quality and rarely hacked. All governments are fully trustworthy, especially around espionage and privacy, and have a perfect track record of never lying to the public.
Look, the European Commission stated that it cannot be hacked; "hackers cannot take control of it", from ec.europa.eu. They built an unhackable device. I am not sure what you could be worried about. If the government tells you something cannot be hacked, then it cannot be hacked. Furthermore, none of the EU member states have been found using other infrastructure to violate privacy laws.
It's not. It just stores the last speed/wheel position/brake state data that it receives when the "collision imminent" condition activates. In some cars this can be literally the same signal that deploys the airbags.
> Look, the European Commission stated that it cannot be hacked
Pretty much. It's just a normal LTE radio, that is normally inactive. It technically is hackable, but I'm not aware of any hacks of baseband firmware of this severity.
And come on. Car manufacturers, which are notorious producers of insecure software, are legally mandated to make an inexpensive device which includes an LTE radio and a connection to the vehicle buses, and you think that is... unhackable? I can't tell if you're trolling me, but your average blackhat only needs 1 of (shitty car OEM software/LTE radio/vehicle bus connected device) to break into a system. This system is a trifecta of hackable crap. To call that, of all devices, "unhackable" is priceless.
The original standard version defined only one location datapoint, the more recent version defines two additional _optional_ points ("recentVehicleLocationN1", "recentVehicleLocationN2"). It also allows specifying the number of passengers.
The mandatory datapoints include the location and direction of the vehicle, but they can be acquired as needed.
> I can't tell if you're trolling me, but your average blackhat only needs 1 of (shitty car OEM software/LTE radio/vehicle bus connected device) to break into a system.
I'm not aware of black hats hacking into a modem that is passively tracking the mobile networks. It's theoretically possible, but I'm not aware of such feats.
The modem does not have to passively track the mobile networks; it can do what it wants. The common OEM implementation these days is that the physical device that does eCall does several things, including eCall, over the same cellular radio. There's nothing stopping the OEM from connecting to a random website and eval()ing the result.
You seem confident in the security of this unhackable system so I will point out some of its other security weaknesses. Several eCall device implementations include Bluetooth modules (both "unused" as part of hardware and implementations that use Bluetooth.) Bluetooth is as secure as a wet cardboard box, so you could take BlueBorne or one of the six million other Bluetooth exploits that work on a non-discoverable device just sitting on the shelf, get in that way and boom, you can transmit whatever you want over the cellular radios. Vehicle infotainment systems are pretty insecure on average and are frequently hacked, so you could take over the infotainment system, get into the CAN bus that way and then send bad data to the eCall system, which is in the business of processing and responding to CAN data.
But those are just a few of the million ways; you could write up attacks all day long and you wouldn't scratch the surface. The facts are: this is a system with cellular radio(s), a CAN bus connection, sensors that constantly listen and interpret data; this is a large attack surface, built by OEMs that write notoriously insecure software. It is, by any reasonable judgment of those facts, a pretty hackable system. And yet, the European Commission goes around telling people that it "cannot" be hacked.
Anyway, this will be my third comment in a row here telling people that their comment is plain incorrect, so I'm going to have to leave the discussion here. I hope that my words provide some food for thought - for the next time that a system that could track you becomes legally mandatory.
I have the standard open, and I don't see it.
> The modem does not have to passively track the mobile networks; it can do what it wants.
Sure. So just choose an automaker that doesn't provide data subscription services and/or don't pay for them. The eCall requirement in itself doesn't require tracking.
It's been a while since I've seen the pro-surveillance argument of "well, you can just do this uncommon or difficult thing if you want to evade surveillance!" In several European markets, for several vehicle types, there _are_ no vehicles without OEM-connected cellular radios. Some OEMs don't even advertise it; it is used for activating features, or "security." Other OEMs will not _fully_ disable cellular connectivity even if you stop paying. If you're an OEM mandated by EU law to include a cellular modem and a location recording device, you might as well make some use of that - and they do.
https://www.slate.auto/en/personalization
Oh, and every year there's "only three days left to invest!"
Such positioning could be what the intended customer base react well to.
For example, mechanical window winders would need a whole extra disengagement or locking mechanism for child proofing.
I’d much rather they included a $200 system, since ~ 100% of their customers will want to be able to have speakers in the doors and a mic in the dash (at the very least).
I never had a manual window winder fail to work, but electric window buttons breaking or the motor getting stuck (e.g. in icy conditions) has happened at some point in every car I've owned.
The convenience factor hugely outweighs the rare failures for me, but I could see why someone buying a Wrangler for its intended purpose might actually prefer the manual option.
How old are you? Back in the 70s-80s these manual ones would break all the damned time. Of course US cars from that age we're commonly crap.
However I do live in the UK so they were European/UK cars - Vauxhall, Volvo, Citroen, Renault.
I have. It jammed. When I tried to release it the glass fell out (into the door).
Electric winders tend to experience motor failure leaving you with a stuck window. They don't stress the mechanism that much.
Manual winders, you would generally experience mechanical failure. If you were lucky, you broke the handle. If you were unlucky, you broke the gears and had the pleasure of watching your window fall into the door.
* Bed size is just five feet
* Towing capacity is just 1000 lbs
* Not AWD
None of these can be retrofitted after the sale.
Where I live, it'd struggle to be called a "truck" with these limitations.
Not everyone wants to spend 40-80k on a bloated luxury-truck-ized F150 when they only need to carry something oversized maybe once a year.
I like the "starts out cheap, then upgrade it later" premise of Slate, and I like that it's electric, but it'll only really be a toy with the limitations I specified.
But if you have even just those once-a-year "need a truck bed" needs the gap between "SUV with fold down seats" and "actual truck" is pretty substantial.
I think the set of truck buyers with either:
* just occasional needs for a bed, without a need to put sheet goods flat or such (if you have that just get a minivan these days ;) )
* a fashion-driven desire compared to a van or SUV vs a practical-driven one
is substantial compared to the set of "needs a professional-grade truck" buyers.
The set of professional-grade buyers hasn't changed much in thirty or forty years, but the former two sets have exploded.
Though yes, I could see it being less important if that's not an issue in your area.
I don’t think the Japanese automakers have squandered anything, yet.
A comparable truck gets 18mpg mixed. At $3/gallon, that’s $0.16 per mile. So, the price premium pays back after 100K miles. That’s comparable to milage driven during a long car loan.
I ignored oil changes, tax breaks on used cars, and picked the form factor where EVs are the least economical.
It’s still basically break-even.
We make a ~180 mile trip roughly once / month and could charge on site as we always stay ~2 nights, though probably slow 120V / 15A charge (aka Level 1). My current car would probably be pushing it, range-wise, but I definitely think for the vast majority of our usage, we could be using only EVs if we got one with a 300+ mile range (based on 100% battery usage.) From what I've read some EVs (like mine) struggle a bit below 15% and start to run in "limp" mode.
Its reputation is that of a brand for people who really like cars, who can appreciate the care put into proper engineering and a wonderful manual transmission; or people with an eye for a "conservative" kind of quality. It's basically the new Volvo, but sportier.
https://www.mazda.com/content/dam/mazda/corporate/mazda-com/...
https://companiesmarketcap.com/mazda/marketcap/
Their expectation is that their sales will be stagnant at best, but probably decline for the foreseeable future.
Mazda maintained their relevance and independence by operating their own center of design, engineering, and manufacturing excellence in Hiroshima, and exporting the results to the rest of the world, since at least the 1960s. As I mentioned, that thread is now broken as far as EVs go, with the Changan JV making EVs for Mazda. China is now producing excellent EVs that surpass the capabilities of ICE cars at a fraction of the cost/price, thanks to continuous improvements in LFP battery technology. China also dominates solar, which (together with the batteries) solves the grid stress issue for large EV deployments in most regions of the world. Together these exports are likely to disrupt Japanese, US, and European ICE exports and energy markets throughout the world, no matter what tariffs the US chooses to enact.
Mazda and the rest of Japanese companies slept on it, led by Toyota's trust in the hydrogen-powered future that didn't materialize, even while Panasonic had the best batteries in the world. The time to invest in these platforms and technologies was 15 years ago - now they will have a far harder time financing this and finding technology development partners. Sure, they can survive - not thrive - on existing ICE exports for a while, but they will face a shrinking market and stronger headwinds - and are likely to lose their independence, which is what allowed them to design great cars. Don't believe me? Look into what's going on with Nissan (which squandered an even bigger lead - the world's first mass-produced EV).
This has been a fantastic decision, as a large number of EV manufacturers have gone bankrupt.
I will never want to listen to the radio. I would love to remove radio as an option. I would love to have no fallback as an option. But no, the car just f-n loves the radio and will not stop trying to force it on me.
Oh yeah, and the radio is buggy and could get stuck if I tune into the wrong station. https://www.bbc.com/news/technology-60333765.
This car definitely tries too hard to be smarter than it is. There's all sorts of exceptions that keep the doors from auto-locking when I walk away, and I would turn all of them off, but I can't. Walk away too fast? doesn't lock. Open the rear? won't auto lock. Car just doesn't feel like it? doesn't auto-lock.
And god forbid you hit the unlock button when the passenger has already unlocked it. Anxious beeps from the car for several solid seconds. That is not an error condition!
Performance and reliability have been great though. They just need to stop trying to be smart. They're not.
https://www.cx90forum.com/threads/fuse-box-diagram.172/
Something foul and malign is afoot at Mazda these days.
[0] https://www.cx90forum.com/posts/2706/
I even adore the scroll wheel and wish it could be in any car I own in future. Yeah it takes slightly longer to do certain actions in CarPlay, but I can do it so much more safely than I could in the Civic I had before. The infotainment boots basically instantly; as you mentioned CarPlay starts itself, and the patronising-but-mandated “don’t use this in motion” warning dismisses itself. In the Civic I would be half way down the road already by the time it booted, blindly prodding at the screen to try to dismiss that warning so I could pause the podcast that started playing itself because I plugged my phone in.
And, while my 2022 car predates the stupid auto-re-enabling ADAS requirement in Europe, the 2024+ models have single button deactivation. I dunno how, cause it’s supposed to require a minimum of two presses legally, but it sure makes me wanna stick with Mazda.
However that makes the upcoming 6E that much more disappointing. They’ve partnered with a Chinese manufacturer, I assume because they don’t have an EV platform of their own ready yet. Looks fantastic from the outside, but the inside is a sea of touch screens with barely a physical control in sight.
So I only use Google maps with Android Auto now, but cannot put the turn by turn display on. Also, who knows what telemetry Mazda is sending home on me without me knowing or wanting them to. Probably selling it to data brokers.
As for selling your data, yes absolutely. It goes to Connected Analytic Services which is an affiliate company of Toyota Insurance. Toyota Insurance Management Solutions (TIMS) is another name to look up. Subaru sells your data to them as well.
When looking at who is doing it right, I wouldn’t put Mazda on a pedestal. They simply are behind the competition.
As an aside a lot of people like to levy criticism on the infotainment screens which I think is very well deserved, but then people text and drive, watch YouTube videos, and do all sorts of crazy things too.
Instead of levying criticism on these distractions (let’s include billboard too) we should instead focus on just reducing car usage since we won’t stop people from being distracted.
The safest car is the one in your garage.
Instead of levying criticism on these distractions (let’s include billboard too) we should instead focus on just reducing car usage since we won’t stop people from being distracted.
This argument to me reads like one for abstinence from sex. The world is not so binary, we can both criticize distractions and build communities where car use is not a necessity. Not to mention in most jurisdictions some of these distractions are criminalized.
Criminalization of texting and driving and such doesn’t matter unless you enforce, and we don’t enforce. So it’s de facto legal. Who cares about infotainment screens at that point?
As for criminalizing texting, I’ve heard enough people getting caught and getting big fines that it works enough for me to dissuade me from doing it.
If you’re focused on less death, sure we can criticize infotainment screens, but the energy is much better spent in demanding enforcement and in whatever we need to do to reduce car usage. Otherwise you’re kind of wasting your time, unfortunately.
https://pictures.dealer.com/s/surprisefordvtg/0292/10a2adba8...
It's like a window into hell.
https://www.ford.com.au/content/ford/au/en_au/home/owners/te...
(Really. They did. No, you can’t adjust the steering wheel position enough to fix the problem.)
Also at least personally I never change the fan speed but just set the temperature I want.
No one likes ads, no one likes their data being collected. The sooner insurance and car companies understand that, the sooner they get out of the maelstrom of false revenue from ad- and spy-ware programs.
The only data I can find relates to Chinese vehicles which shows some concerns, but that's understandable given they are built by a foreign adversary.
https://www.autopacific.com/autopacific-insights/2024/5/22/y...
What percent of users understand how much data is being collected about them?
The thing is that car manufacturers have been fucking up software in cars since... forever. The second car play and android auto hit the scene, that's all anyone wanted.
There's more benefits than just what's on the surface, too. Even if the car software is perfect, it doesn't have access to the same data your phone does. It won't put your contacts in your navigation, for instance.
Look, can car makers make somewhat decent software? Probably, if they burn enough money. But is it even worth it? I don't think so. People already use their phone hours a day, just let them use that.
Changing lock, light, and anudio (bass/treble/sub/fade) options. Map integration with fuel capacity (they only recently do this for EVs). Checking service intervals, recalls, etc.
If CarPlay had APIs/toolkit to serve those functions, it could 100% replace the UI that the manufacturer delivers (and nobody likes).
Tracking, phoning home (with related privacy issues), etc:
* https://arstechnica.com/cars/2024/09/flaw-in-kia-web-portal-...
* https://arstechnica.com/information-technology/2019/10/five-...
There's zero chance a car manufacturer is going to nuke some of the most desired features of modern automobiles for some undefined cohort of privacy conscious consumers.
Most younger drivers would even buy Chinese vehicles despite their privacy concerns.
https://www.autopacific.com/autopacific-insights/2024/5/22/y...
It just should be said, that all these features could perfectly be implemented without violating privacy. You just have to use another system not from Apple and that other advertising company.
I have an iPhone with no social media apps. Mainly use it to check e-mail, Maps, weather, and SMS/RCS.
I feel confident that there is minimal third-party tracking, and that Apple themselves are relatively honest.
> support wireless CarPlay and android auto
Removing LTE doesn't cost me real-time traffic updates because (preferred maps app) is running on my phone which already has LTE. Streaming media? The media is being played from my phone or streamed via my phone, which already has LTE. I'm not sure what "remote controls" are in this context? Letting me set the A/C fan to high from Internet (almost certainly via a browser or app running on... wait for it... my phone)?
We've already paid for the LTE modems and app integration on the phone side of things, don't need to pay for it a second time on the car side or have to deal with the vehicle manufacturer's terrible implementations of navigation apps and media streaming services or yet another vendor collecting telemetry about me and reselling it to whoever wants to pay.
If you're gonna build that crap in at least go back to a standard-sized replacable module.
You are complying by installing it, the customers are the ones [easily] removing it [because you were a bro].
Those safety add-ons are there for a reason.
If a car is connected to the internet, it does not automatically mean that it is also collecting car data and sending it somewhere.
Secondly, the government makes both regulations (cell tower data retention, emergency calling). So in combination the government regulated user tracking. Where exactly it happens isn't really that important. Once the data is being collected it'll end up being used. Usually it starts with major crimes and anonymized statistics, once the automated infrastructure has been created it'll be abused either illegally by police overreach or legally by a tough-on-X government changing the rules. It's not like this is some novel pattern...
In fact, I'm going to find mine and faraday that shit this weekend.
It's safer and convenient to have a connected car. I like it that way. When you can open your car app and check the location where it is parked, amount of gas or request ventilation/ac.
I am happy to share my address and phone number etc. but not to be tracked wherever I go or have everything I do on my computer or within my home monitored. These are different things.
what is the killer app of a connected car? businesses might want to watch their fleet but does anyone else care
It requires at least a basic cellular module for eCall in Europe since 2018, so car manufacturers use the already present hardware to provide more services. Maps and updates (live traffic view), internet hotspots for passengers (IIRC, Tesla does that one), entertainment that doesn't rely on a phone, firmware updates, feedback of driving data to insurances (yes, some insurances offer discounts in exchange for proving you "drive safely"), position data for leased/financed cars in case they need to be repo'd, synchronizing stuff like seat and mirror position across a fleet, remote pre-heating, "put packages in my trunk" access for parcel deliveries to thwart porch pirates, uploading data from real-world traffic situations to train AIs (again, Tesla does that one)...
There's quite the laundry list of nifty to nasty things that can be done with a connected car.
* giving me the current fuel and battery levels in the app
* giving me an ETA on when charging is finished
* locating my car
* telling me if the car has been sitting there for a few minutes with ignition off but doors unlocked, giving me the option to lock them remotely
* telling me about open windows, giving me the option to close them remotely
None of them is really crucial, but for a hybrid or EV, getting the ETA for when the charge is finished is pretty useful.
When is this actually useful? In the ~12 years I’ve been driving, I’ve never needed to know the fuel level of a car when I’m not in it. I guess maybe if I’m planning a road trip and need to know if I’m going to have to stop for gas before I leave? But I’ll figure that out when I get in to leave and I’m probably not leaving with <10 minutes of margin.
> locating my car
Again, never once have I not known where my car was. I think my phone keeps track of where I park too already? But I’ve never needed that feature. I guess if it’s stolen and the thieves don’t know how to disable this, it could potentially be useful for insurance/police.
> telling me about open windows, giving me the option to close them remotely
This could be useful. I’ve never left windows open by accident before, but I have left them open on purpose - if there were an automatic notification when this happens, I’d probably just eventually turn it off to reduce the irritation from false positives, and then not be notified if I ever left them open by accident.
> remote door un/locking
I had a Lincoln that had this feature, while I was working as a reverse engineer/pentester. Took me ~45 minutes to be able to send an unlock request to the car, unauthenticated, and have it open the doors, over the internet. Pretty sure that’s never been fixed (at least, it hadn’t been when I got rid of the car - model year 2016, which was identical to the 2013s, and I got rid of it in 2022). Needless to say, not a fan of that kind of “feature.”
I could see charging ETA being useful if multiple people are using the same car and for whatever reason can’t communicate that sort of thing with each other, and don’t have a feel for how long charging takes. (I’ve never owned an EV, but I imagine that you plug it in when you get home, and then it’s ready for you in the morning, so I don’t really know what the use case for knowing the ETA is in that case. Maybe if you’ve been driving around all day and need to make a long drive in the evening? I still assume you’d know how long it’ll take to charge when you plug it in though. And if you're at a fast charger, don’t they have a screen that gives you the ETA when you plug it in? I’ve only used one before, but it did that, and it was accurate to within 30 seconds, so I’m not too sure how useful it would be to have the ETA on your phone in that case either.)
* charging stations have different powers
* charging time depends non-linearly on the remaining change
* ... and it's also temperature dependent (though only a little, with my plug-in hybrid)
I cannot plug in at home, but there's a public charging station around the corner, 3 minutes walk.
So I arrive there, plugin in, and the car gives me an estimation when charge might be finished. The initial estimate is sometimes off by up to 30 minutes (usually less). Sometimes I also forget the estimate, because I'm too busy with other things.
Getting a notification when the charge has finished, and an updated ETA on demand, is a notable QoL improvement.
It feels a bit similar to bluetooth headphones: I never complained about the cable before I switched to bluetooth. But now, I'd find it annoying to go back to cables.
Eventually I expect cars, chargers, calendars, and the electric company will somehow integrate so that you can plug in and the system figures out when to charge your car. that is a complex project though with privacy concerns that are hard to address.
https://www.wired.com/story/subaru-location-tracking-vulnera...
There are several reasons to have internet connectivity in a car. For example, you might want to start your car remotely (e.g. winter time and you want to pre-heat it) or you'd like your onboard navigation maps to update automatically, or you'd like the latest traffic reports (if available in your region), yadda yadda.
While there are a lot of people that love Android Auto and Apple CarPlay, there still remains a sizeable group of people that want to have a navigation solution without using their phone, or to be able to enjoy their car without having a smartphone at all.
Yes, Tesla has one of the best user interfaces in a car, and has set the bar high. But just because they have OTA updates it's now called a "Software Defined Vehicle"?
See Rivian's intro on their ECU design and Zonal architecture: https://youtu.be/6ZBko4TvfJY?t=137&si=-SKL_iFqZFnHE8nQ
This might sound like purely implementation detail, but having the (non-safety-critical) "business logic" of a car as software gives the manufacturer flexibility to late-bind behavior as new use cases / demands inevitably get discovered.
Something can simultaneously be a good idea, buzzword'd by marketing, and/or deviate from the original intentions.
Tier 1 suppliers have enough resources in both know-how and manpower that I have been wondering if they could do a platform car. Provide a basic frame that passes crash, provide a basic engine that passes emissions, provide basic safety, etcetera.
Then invite other parties to upgrade components. Package lots of air between components to simplify compatibility.
I suppose the only way to get this going in the real world is a big military contract, but I am wondering if it wouldn't be smart play for everyone involved. It would be deadly for a bunch of traditional automakers, but they can't do anything preventing it.
Come up with few general hardware modules, enough to replace the head unit, body controllers, ECU, climate control, and ideally driving automation, and software to run them. Everything minus safety modules like the airbag controllers, and then license them under Fair/non-discriminatory terms.
Then, a variety of automakers get access to core functionality and cheaper hardware to run it. That means that the cars themselves can have higher quality software, cheaper hardware (from cutting out companies like Bosch that charge exorbitantly for things like a windshield wiper controller), and thus deliver more value to customers.
Is this not just Android Automotive? A lot of Volvos use it, it’s a lower-level OS type thing that sits below Android Auto or CarPlay.
And past HN discussion https://news.ycombinator.com/item?id=32447650
https://openinverter.org/ https://youtube.com/@evbmw https://news.ycombinator.com/item?id=43898280
I'd like a car with zero screens, no internet connectivity possible, and maybe one audio input and a radio.
Also I drive a manual, which here in the US seems to be almost unheard of.
As an aside, what's next? You can't buy a chef's knife without wifi?
It's uncommon but some enthusiasts still drive them. My last two vehicles have been manuals. Planning to keep driving them as long as I can. 8)
All of my last 5, including my current vehicle are manuals. Almost impossible to find and a dying breed.
This was something that really hit me when the Internet allowed game developers to ship a game that wasn't done. You got the game, and the first thing you did was download a "patch" that was at least as big as the CD the game came on (several hundred MB). I've got "released" Windows98 games on CD that are essentially unplayable because what was shipped on the CD was unplayable and without the update server on the network sending out those critical fixes, its never gonna work. For game archivists that means finding a fully patched install and then preserving that.
This is a shitty experience that serves manufacturers but not their customers. I don't expect it to get better any time soon but I wish it would.
Vertically integrated companies do this very differently. Tesla pioneered this. The Chinese copied this and at this point you also have companies like Rivian and a few of the legacy manufacturers that are doing the same. Effectively they in house all the software and e.g. Rivian runs the software on a handful of hardware subsystems instead of having hundreds of chips with their own firmware for things like the wind screen wipers, the software that controls the windows, the AC, the keyfob, AI driving features, and so on.
I mention Rivian here because they just did a deal with VW to start doing the same for them.
The issues here are not just technical but cultural. I used to work in Nokia when it was in the (slow) process of figuring out that they were a software company rather than a hardware company. Then Apple and Google came along and they were slow to adapt their internal processes and management. Apple makes firmware that goes on their phone. They provide OTA updates. There's only one supported version of that firmware: the current & latest one. It's the same for all phones they still support with updates. Nokia did the opposite. They forked their software for each product variant (dozens per year). And they did not do OTA upgrades. So most of their phones weren't updated at all (by users), and would typically ship with bugs that had already been fixed on other branches of the software. And it would ship on the schedule of the manufacturing process, regardless of the state of the software. With all the obvious consequences. Nokia got a well deserved reputation of shipping half baked software.
By the time MS bought them out, they had learned and improved a lot but Apple and Google were running circles around them by then and it did not matter anymore.
You see the same with car manufacturers currently. It's all about the buttons and the bling. They have a gazillion of upsells, features, special trims, and what not. And it all adds up to a whole lot of nothing if the software experience isn't great. That's why VW is paying billions to Rivian to fix that for them.
Their cars are too expensive, have too many chips and wires, and their software just isn't good enough. And they don't have ten years to figure this out for themselves. That's what Rivian is supposedly fixing for them.
The only formula I know that works is "hire good people and listen to them". From experience, the only way legacy companies can do this is acquire and/or seriously partner with companies that have established a track record in what you need (even if it is only a couple of years, as long as they are _delivering product_).
As software effectiveness/innovation speed/productivity continue to increasingly crush legacy industries, it is extraordinarily frustrating to see how hard it is to make (seemingly simple!) changes.
p.s.: nice to see you Jilles! :-)
Despite the problem having the hallmarks of a backend issue (many cars with the same software running into the same issue on the same week), corporate is still insisting that it's a hardware issue and trying to sell us on $5k hardware replacements. I love the car for its build quality, but almost kind of wish I'd gotten a Tesla given how bad VW is at software.
I could probably whip him up something nicer if only there was just a Nuc or something in there somewhere.
I would also guess (completely un-informedly) that because the simple (and probably correct) answer is very difficult, a lot of companies are trying to avoid it by doing things that are more complicated but also easier. And because they are more complicated, it is not immediately obvious why they won't work....but they won't. Which is resulting in the repeated failures.
Yep, I can see it.
These systems also have to work correctly (100% of the time) in a range of conditions and it needs to not drain the battery while the car isn't being used. They also need to start up quick on relatively low end hardware. The car creates a very hostile environment generally for electronics. There is lots of dirt, muck etc that will literally work its way in everywhere. There is also a bunch of regulations that have to work almost internationally.
Car companies are not software companies. If you are a software developer not in a software company, things are much more difficult as the organisation just isn't geared to deal with software development generally. Combine this with it being a massively complicated product (modern vehicles are complicated) you are setting yourself up for failure.
No. I don't want it. I want Not to have it.
I don't want a touchscreen. I don't want a computer car. And I definitely don't want an internet-connected car.
Apple had a secret test track in Arizona, with buildings made from shipping containers. You can see it on Google Maps under "Chrysler Oval Track".