Hacker News Reader

Comment by nimbius | Hacker News Reader

As an automotive mechanic, this is awesome. You have no idea how many sensors/controllers in a car can only be reset by violating the DRM.

Example: the suspension control computer for newer fords (so far just raptor and newer commercial F series) is separate from the ECU, and if it faults out you need to buy a new $2500 computer and sensor pack. You can, however, replace a commonly blown diode or fuse on the computer but opening the case causes the device to stay "in fault/service"

With a raspberry pi and a bit of python however you can reset this tamper code once the whole assembly is closed. but since it "violates the DRM" most shops generally just bill for the part and labor.

View on HN

> With a raspberry pi and a bit of python however you can reset this tamper code once the whole assembly is closed.

This sentence is interesting to me. It tells me that now mechanics need to learn programming skills to be successful. Software really is eating the world...

I'm convinced most successful mechanics could be successful programmers. The skill set of looking at a problem and debugging until you really understand the architecture and design of what you're looking at is very similar in both fields.

American McGee is probably the most famous example of this.

> Dead-end jobs followed – washing dishes, polishing rings at a jewelry store, ringing orders at a record store and, finally, fixing Volkswagens at Uncle J.O.'s auto repair shop.

> The tinkering that's required of mechanics appealed to the teenage hacker. The $6-an-hour job "was the first time I ever felt like I accomplished something at work," McGee said.

> On breaks, covered in grease, he'd write inventory and customer tracking applications for the shop.

https://www.wired.com/2000/12/the-great-american-mcgee-game/

A lot of software engineers around me love poking in their cars

I'm a software engineer and I do most of my own car work. Can't do everything though, I don't have a lot of room for tools that I'd rarely use.

Paid someone to fix my air conditioning vents though because I'd have to half disassemble the interior of my car to do it and I knew it would take me longer to do than I could have the car out of commission. I really hate the vacuum controlled vents in mid 2000's Ford trucks. I had a vacuum leak - somewhere.

Motorcycles are great for that. Except for a few really high end modern bikes (mostly BMWs) they are mostly repairable with nothing more than common tools and a little ingenuity.

I really should just be riding my bicycle to work...

Before I bought my first computer, at 16 - 22 I was repairing and thus disassembling and assembling back my motorcycles and cars on a monthly basis. I didn't have enough money to buy a good working one so I had to tinker with it very often.

The most extreme I did was to replace the car body after an accident I had (slippery road) in my own garage. After a week or something I got a running car again, with a different color and car body number but whatever ^^.

My nerdiest friend and myself have done engine swaps in our driveway, ranging from LT1s to VG30DETTs. Once we got the grease off, it was Wow time!

I'm not interested in cars, but I built my own house. There are plenty of program-a-like jobs and hobbies to go around.

Are they not afraid of regressions?

With car it much easier to provide a decent test coverage though :)

I'm convinced most successful programmers can be successful mechanics for the same reason!

Most programmers I know tend to be handy in some way, whether that's fixing their own cars (myself), wood working, or some other skill which requires them to think through problems and work with their hands.

I actually view fixing cars as not only a hobby of mine, but as a fall back career should I ever become bored of software or need some quick cash.

> Most programmers I know tend to be handy in some way, whether that's fixing their own cars (myself), wood working, or some other skill which requires them to think through problems and work with their hands.

I can relate to that. In fact, I find it downright relaxing being able to work with wood (I like to restore mid-century furniture as a hobby) and compared to all those digital, virtual tasks and projects it's so refreshing being able to physically touch, feel, see and use the result of one's work.

My brother is a mechanic and both worlds have so many similarities.

We joke that the big difference is that I got AC during the summer and he doesn't.

That's what I thought, and then I fragged a Ferrari engine. Dunning-Kruger is not a river in Egypt...

I leveraged my embedded systems knowledge into a cottage side business rebuilding the old Magneti Marelli ignition modules in 308/328 models with new ARM Cortex based electronics. There's nothing magic about Ferrari engines.

It's funny, I've met a lot of people that thought dsp and algorithms engineering to be magic and yet I find those fields to be a straightforward and fun. On the other hand I have a year of professional experience in ARM Cortex M4 programming and I find it to be magic that any somewhat complex embedded system works. Flip one bit the wrong way and the whole fracking system won't even boot. It took me six months and a significant chunk of my professional reputation to get some biomedical algorithms to interface with offline data on a fat16 formatted USB device.

> I find it to be magic that any somewhat complex embedded system works. Flip one bit the wrong way and the whole fracking system won't even boot

Absolutely agreed. I think "what if this fails?" a lot in situations that most likely really don't require it.

Had to look this up. Denial, is a river in egypt. But thats just a meaningless pun. The Nile is a river in egypt. What do you mean by your saying?

I think it's just supposed to carry the same meaning as "Denial is not a river in Egypt," which is: hey, denial is a real thing, and you're facing it right now.

I'm not the OP but I hope that helps.

Dunning-Kruger Syndrome is basically a manifestation of denial that you suck at something.

Programming prepares like almost nothing else to pay attention to details and to actually read and follow directions, it's like surgery in that regard one little distention and the programs malfunction or die.

This translates very well I in most task oriented jobs, like, maybe your average programmer won't be able to invent a new gourmet dish but I expect most of them to be very competent cooks with a recipe on their hands.

I am commonly told I am a good cook when I make stuff and I just say I follow the instructions. I guess when you spend a lot of time writing instructions you come to appreciate when people write them for you.

For me, if this software developing thing doesn't work out, my dream job would be to purchase, restore and re-sell classic cars. I already love tinkering with them, but there is only so much I can do with a full time job, and living in a flat.

I grew up working on cars and motorcycles. The mindset is very similar. Since I work more in DevOps than as a developer, I often think of myself as a software mechanic.

Interestingly, I find the musician mindset is also very similar, particularly musicians who compose or improvise.

Software Mechanic is a pretty good term. A lot of people out there claim to be software engineers (creatively design solutions to problems), when really they are mechanics (apply technical skill and knowledge to already solved problems). Web development is an example of a largely solved problem, the answer to every problem in webdev is to look it up in the manual (Google it). Adoption of this term would clear up a point of mass-confusion in our industry.

Exactly, my debugging is so much stronger than my architecture skills that I've been calling myself a software mechanic for a while as well!

Software Mechanic is a great brand. I'm stealing it :)

"Software Janitor" is my personal favorite. I should probably carry around a bag of pencil shavings on principle.

Yeah, I used to say CTO = main janitor of tech.

Software mechanic says: I don't mind getting my hands dirty while doing intricate work (computers, who understands 'em?). Janitorial work seems to be regarded as lowly drudge work, even though being a good caretaker/warden means preventing problems from occurring in the first place, like watering the garden and keeping leaks in check.

Everyone is trying to keep a cupboard in order somewhere :).

In my enterprise dev mgr role in maintaining a legacy codebase while introducing new features I used to talk w/ my team in terms of "software gardening".

Most mechanics are hardware hackers. Instead of searching for the problem on stackexchange/stackoverflow, they search service bulletins and automotive message boards.

I'm sure that the software in modern cars causes all manner of stupid problems that never existed before, and could be easily eliminated by giving the mechanics access to the source.

Its almost as if you could call programming 'software engineering'

As someone that grew up with mechanical stuff and ended up in software I agree :)

It tells me that I really shouldn't buy a Ford car, since apparently billing for repair parts is more important to them than building a robust and repairable car. Adding a tamper resistant circuit to random bits of your car's electronics is adding cost while reducing robustness - who wants that in their car?

I hate to break it to you, but that's about every manufacturer now. Imagine how it's going to be once "drive-by-wire" becomes the standard. I'm right there with you though, wishing stuff stayed a little more repairable and maintainable.

Ford is basically stopping production of cars anyway as they found that trucks are more profitable. But yeah, don't buy one of their trucks either. But what manufacturer is an alternative? Tesla set up DRM to artificially limit the driving range of their cars until you pay them a ransom to simply let your car drive longer distances!

> Tesla set up DRM to artificially limit the driving range of their cars until you pay them a ransom to simply let your car drive longer distances!

I think the effect of this sort of DRM is underestimated in many conversations about post-scarcity.

People don't seem to appreciate that cryptography will create programmatic scarcity where previously we only had scarcity due to physical laws.

The bright side is that it's easier to crack software than it is to find new mineral ore deposits

In the past tesla sold the same hardware & number of batteries in two car models, but saved 10k in price for the shorter range one; and you could upgrade to the longer range by paying more. Just like for CPUs. In the new mid-range m 3 they have fewer batteries than the long range. Tesla eventually stopped selling the software upgradeable one, maybe because they bugged people.

> Tesla set up DRM to artificially limit the driving range of their cars until you pay them a ransom to simply let your car drive longer distances!

That's a lie. Yes they have DRM in place to artificially limit the range of certain models, everything else you said is mischaracterization, misrepresentation, and lies.

They saw opportunity in offering different ranged batteries of the same model, however the cost to produce two different battery capacities was too high. Instead of hiding this fact from the consumer, they instead offer it as an upgrade path if you one day have a need for the additional range.

This sort of thing is done all of the time in manufacturing to streamline costs. They even do it with their Autopilot system, all Model 3s come with the same sensor package but the feature itself is unavailable unless you pay to unlock it.

There's an advantage to doing this in that you do not have to guess at what consumer demand will be because you can lock/unlock features as needed in software based on consumer demand.

>(parent) That's a lie

OK let's see where is this lie...

> (grandparent): Tesla set up DRM to artificially limit the driving range

> (parent): Yes they have DRM in place to artificially limit the range of certain models

OK no lie here. So it must be in the second part of GP's comment?

> (gp): until you pay them a ransom to simply let your car drive longer distances

> (p): they instead offer it as an upgrade path if you one day have a need for the additional range

So you just confirmed everything from GP's comment, where are the "mischaracterization, misrepresentation, and lies" you speak of?

The word ransom.

I think it's more accurate (certainly more charitable) to call his statement hyperbole. Calling it a ransom seems like an exaggeration but not really a lie. But if you accuse him of lying when he's not then you are lying :)

Ransom is when someone takes from you and only returns what they took when you meet their demands.

Tesla didn't take anything from the buyers of the car owners who paid for 60 kWh capacity. It gave them what they wanted, it just happened to delivery it in a vessel that could hold 75 kWh.

Rather, the car buyers were offered the option to purchase the exact same car at a reduced price because it was artificially limited to 60 kWh.

I'm not clear what people are upset about. Are they saying Tesla should have made 60 kWh packs or sold the 75 kWh packs at two prices and hoped most people would opt to pay more money?

In the former scenario an owner of a 60 kWh car would have to replace the entire battery to upgrade or trade in their car to a higher capacity if they found they needed it. In either case their costs would be higher than just paying to unlock the restricted capacity.

In the later scenario Tesla loses money because no one opts for pay them more money for the same product.

The ransom part is when someone says that you have to pay them in order for them to release something that otherwise costs them nothing to release (you're not paying them for work or for resources that they want for themselves).

In some cases, someone takes something from you and then offers to return it for a ransom. That sentence wouldn't make sense if the entire thing, including the initial taking was itself "ransom". Ransom is just the second part.

People are rightly upset at the very concept of artificial limitations. The more stark it is that a price has no connection to costs but merely to a marketing game, the more offensive it is.

I'm not saying the answers are easy, but there's a level of transparency and honesty we can rightly expect in our system.

So, if to produce products economically and satisfy market demand, we find we must rely on some people paying extra money and others paying less (because such a setup allows the low price to be lower and thus to reach more customers), then we can flat out say it. We can say, "this is the same product, same features, same cost to the company, but by artificially limiting the features, we can get some people to pay extra and then pass on some of that as a reduced starting price".

At least then people won't feel like they are being marketed to dishonestly. Of course, people can see reasons to object to the idea of artificial limitations, such as how they privilege the wealthy in ways unrelated to actually needing to use their wealth to cover additional costs (it's different to have rich people have extra software features than for rich people to have more land or other actually scarce resources).

Anyway, "ransom" is indeed a framing, but it's not crazy, it's reasonably honest. For example, people describe running a crowdfunding campaign to pay for existing software getting freed under an open-source license as a "ransom".

We could accept that Tesla's ransom approach is a good solution. Or we could suggest something else (maybe some sliding-scale pricing?)

In general, prices in our market are screwy and people intuitively get that. The more screwy or manipulated or dishonest they seem, the more annoyed and distrusting people feel.

Yes, people would feel different about literally getting cars with physical differences, but you're right that purposely producing a worse product at very-little cost savings just to hit the lower price point isn't actually better in the big picture.

So what you're saying is that it's ransom because it's functionality you want but didn't pay for the right to use?

And that a company should either not sell a product at all, or only sell it completely without restrictions or limits?

Otherwise it's dishonest and ransom?

If you hire a man to mow your lawn do you have a right to be upset that he refuses to clean your septic tank for the price you paid him to mow your yard? You are after all paying for labor from a tool that's capable of doing more than just cutting grass.

Seems like a stretch to (repeatedly) characterize your parent's post as a "lie". You're both characterizing the same practice in different ways. I'm inclined to agree with everything else you said (although I dislike the barriers to independent maintenance of Teslas in general), but the parent's post was a spin, not a lie.

Calling it ransom isn't spin, it's disingenuous.

They paid for x range and received x range. They were given a standing offer to increase that range for a cost significantly less than the cost of replacing or physically upgrading the battery.

Now if they had paid for 120kwh batteries and received 120kwh batteries that would only deliver 100kwh of range then we'd be having and entirely different conversation.

For what it's worth, I disagree with you but don't think you should be down voted for this reply, since you're explaining your position clearly.

FWIW, if HN would offer a disagree and flag button as two options, we would have far better conversations here.

Flag / downvote is essential, or we would just get spam or become Reddit. But currently there is no way to disagree without punishing.

As it is, many are afraid to post anything other than echo-chamber because they might get downvoted.

Yeah, it's an idea. Either that or they could more enthusiastically push the idea that if you disagree you should explain why, rather than down voting.

Generally if someone makes a solid argument or valid points, regardless of whether or not I agree with them then I upvote them. If I disagree I try to reply and explain my perspective.

I reserve downvotes for absurd or baseless remarks, attacks, and other noise.

People like to downvote off topic or sarcastic remarks, I generally ignore them (neither upvote/downvote) because I think making light of topics isn't necessarily a bad thing and can engage people in a discussion.

* I say generally because no one is perfect and I don't strictly adhere to the above.

That's pretty much exactly how I operate as well. I think it would be a good thing if HN tried to encourage this behaviour, perhaps by reinforcing the initial guidelines on signup with things like reminders when down-voting controversial comments.

We all know the theory of down voting and practice of dwon voting are quite distinct by now.

But they did. They paid for and received larger capacity batteries in their vehicles that are artificially hampered to a lower capacity.

They also stopped doing this practice, the new mid range m 3 have fewer batteries and are not software upgradeable. Is this better? Not sure.

Net-net probably? Less batteries, lower weight, better performance, lower environmental impact from the waste of mining and throwing away that extra lithium you didn't pay to unlock...

If you buy a 4 piece of chicken nuggets at McDonalds and they're out of 4 piece boxes so they put your 4 nuggets in a 9 piece box for you, have they cheated you out of 5 nuggets? No.

Then why, if you pay for 100 kWh capacity, and they give it to you in a battery capable of 120 kWh capacity are they suddenly cheating you?

Did you pay for the higher capacity? No.

Did you get the capacity you paid for? Yes.

Are you able to buy up to the higher capacity? Yes.

They put 6 nuggets in the 4 nugget box.

You paid to eat 4.

Those other two cost a little extra to eat.

You can't eat them, until you pay for a catalyst that renders them edible.

That's one way to look at it but it belies the fact that the box was designed and intended for 6, not 4. They didn't force and extra 2 in the box, it's not out of spec.

Actually, the box is either 4 or 10, but just a quibble.

I do not care that Tesla shipped more battery personally. Just got sucked into the analogy here.

I actually looked on the McDonald's website but couldn't find anything but the 4 pack. I remember working their in the 90s and they had at least 4 different sizes (like 4, 6, 9, 20).

Mine has 4 and 10.

Source: granddaughter orders nuggets any chance she gets

I don't have any problem with them giving me extra capacity if they are unable to deliver what I paid for. What I do object to is them giving me extra capacity, but trying to prevent me from using it. And then claiming it is illegal for me to take a hacksaw to their lock they left on my car.

It’s a purely artificial limitation. You’re not getting anything new when paying for the upgrade, just effectively flipping a boolean.

Yes. And the person paid for 60 kWh of battery capacity. They didn't pay for 75 kWh of battery capacity. The true capacity of the battery is irrelevant so long as it delivers the capacity the person paid for.

Tesla didn't sell the vehicle as having a different curb weight, or a specified number of 18650 cells contained in the battery pack. They sold it based on having a 60 kWh capacity which it does.

Why do you feel entitled to the additional 15 kWh?

It's not about entitlement, it's about the fact it's there. It's about how it feels. It's different when someone sells you something that does 2 things, then tells you that unless you pay them more it'll only do one, forever. It's a nagging feeling. It's mine now, not theirs. Then they ask for tons of money, don't give you anything new or different, just enable the product to do what it could already do -- what I "already paid for." Cost to deliver: $0. Charge: $$$. It's not rational, but it's how people behave and think.

You're not getting downvoted because you're wrong. We know you're right. But you're telling us our feelings are wrong, and people don't take that well haha. We all agree on the facts.

I'm not telling you that your feelings are wrong. Saying "I don't like that Tesla sold a 75 kWh capable car as a 60 kWh car" is perfectly acceptable. Saying "Tesla sold me a 75 kWh car that performs like a 60 kWh car" is also valid though misleading because it belies the truth that the buyer wanted a 60 kWh car.

Calling it ransom or extortion isn't just misleading, it's an outright lie.

You're free to feel what you want about the situation and express those feels but I feel that it's harmful and toxic to then make false assertions because you feel bad about a situation.

Its the opposite of the hacking spirit where you make something do more than it is supposed to. Something many people here take an interest in - its in the name Hacker news.

That's probably the most accurate way to describe it, nice work. You get an updoot.

> They sold it based on having a 60 kWh capacity which it does.

The thing is, it was their decision to give a battery capable of 75 kWh, not mine. And it's OK that they have software to limit it to 60 kWh. But it's not OK (obviously in my opinion) for them to prevent me from modifying the software to bump it up to 75kWh. I bought the car, I should be able to do what I like to it.

I actually don't really even have a problem with them including some sort of DRMish thing to prevent me from changing the software, but it shouldn't be illegal for me to try to circumvent it. Again, I bought the car, it's mine now, I should be able to do what I like with it.

I understand that Tesla didn't make the IP laws. What bothers me is that they use those laws in this manner.

Because the fact that they can do such a thing at a profit means the market is inefficient.

Imagine that instead of having a single Tesla, there were two identical companies each selling half of what the actual Tesla does.

Now one of them could offer an unlocked 75kWh version at $1 more than the locked-to-60kWh version (since they cost the same to make and they are profitable) and would capture all the market, so the cost of those two products should be the same.

So Tesla essentially can only do this because they have a monopoly on Tesla-or-functionally-identical cars.

Tesla isn’t really profitable right now, is it?

> This sort of thing is done all of the time in manufacturing to streamline costs.

A good example is Intel and chip manufacturers locking their cpus to a specific frequency when they are capable of higher so they don't have to manufacture the same cpu with different frequencies.

That's a bit different, right? CPUs are binned based on their ability to perform at higher frequencies / not be defective. They're turned into different 'products' so that less-than-ideal parts can still be sold instead of thrown out. 4-core CPUs may be 6-core CPUs where 1 or 2 cores are physically defective pieces of silicon. Sometimes, especially as a process matures, there's too few defective parts, so they take some of the higher performing parts out of the good bin and throw them in with the lower performance ones to meet demand.

This seems more wasteful because you have to mine extra lithium, manufacture extra batteries, install them, haul them around -- and this makes your car perform worse. All on the off chance someone later decides to buy up to the extra-power version?

One's a great way to use extra parts that would have been scrapped. The other's just wasteful...

Intel and AMD have sold hardware as lower SKUs than binned in order to meet demand too. Intel’s also sold upgrades [1] that unlocked features that were already in the CPU.

I guess the closer analogy to CPU binning would be if they found a fault in a P100’s batteries and sold it as a P85 instead of replacing the battery. CPUs aren’t a great analogy though, since the unused material in a Xeon-E that failed binning and became an i3 is pretty minimal.

1: https://en.wikipedia.org/wiki/Intel_Upgrade_Service

also battery packs have nom trivial weight that affects both performance and range.

And the car was sold by weight and battery capacity. No one was mislead about the curb weight were they?

> That's a bit different, right?

Sort of... sometimes CPU manufacturing processes are too efficient or mature and they produce more higher spec parts. They don't then lower the price of higher spec parts, instead they under clock the CPUs and lock them so they can't be clocked to their higher spec.

People caught wind of this fact and began overclocking their CPUs beyond their rate specifications. This started an arms race between CPU manufacturers and buyers. Intel at one point was, and possibly still is, laser cutting the CPU PCBs to physically prevent changing the clock multipliers after they're set.

Cobalt is likely more controversial the lithium.

What you wrote basically confirms what GP said, you are just judging it differently.

The behavior isn't ransom. The customer didn't buy a 100kwh battery and only received 80kwh. They never had a 300 mile range and they didn't suddenly lose 50 miles of range unless they paid extra.

They received the performance they paid for and the option to increase that performance for a cost less than full replacement of the battery.

For some reason I can appreciate the adaptive cruise control being an optional extra, as they've had to build and maintain the software for it.

The battery / range thing no so much.

If half your potential customers are unwilling to pay for a higher range option but will purchase a lower range model for a reduced price that still nets a profit for the same fixed battery cost, what do you do?

Offer a lower range option.

If the cost of retooling a line to make 2 battery types, or setting up a new line from scratch results in a greater cost per unit, regardless of capacity, than 1 line producing the single high capacity battery then why make two battery types?

Are you saying they should just offer only the higher capacity battery at a lower price or just tell the lower range customers to fuck off with their money?

What if you could satisfy both potential customers and offer the lower range customers the option to upgrade in the future at a cost lower than replacement of the whole battery?

What if you could also offer them a good will gesture of added range if they're fleeing natural disasters?

Is that ransom?

Would you be outraged to discover your printer's print speed was artificially capped to not erode the market for a faster model?

Or that your CPU wasn't running at it's maximum potential?

Does it only become outrageous or random when the manufacturer offers you the option to unlock it's full potential?

What about Labor? If you pay a yard service $50 to clean your property and they do it in 4 hours but your neighbor, with a 2x larger yard pays $75 and they still complete the work in 4 hours. Do you demand a lower rate? Faster service? Both?

The fact of the matter is if you paid for 100kwh battery and they give you a 120kwh battery that will only put out 100kwh are you being cheated or held at ransom? Or are you just getting what you paid for?

Why are you entitled to get anything more than what you agree to buy?

I upvoted this comment, just so you know. You've made some good points, food for thought for me.

I agree with his points - logically.

That doesn't make me less upset that my chip has wasted potential artificially locked away.

Sometimes, how you treat your customers doesn't have to be unfair for it to make a negative impact on customer vibe/relations.

I don't understand how their customers were mistreated.

Tesla produced two identical cars that cost exactly the same to manufacture. They offered one at a cheaper price but with a restriction on the battery capacity. They sacrificed their product margin to offer customers a vehicle they wanted or could afford.

They also said "hey if you want the additional capacity you can pay the difference to unlock it."

I upvoted you. And them.

Good, solid discussion. Nice to see.

Every company does this. It reminds me of Casio and their calculators. You can "hack" some of them just by painting a wire with a pencil, and it unlocks features of more advanced ones.

How would you feel about a regular petrol car having extended range as an option, but you knew the shorter range model had exactly the same tank but the fuel gauge was calibrated differently, and the fuel pump stopped pumping before the tank was really empty.

The differentiator the consumer is paying for is range, not # of cells in the battery.

If you buy a 2ghz CPU and it turns out it can run at 2.5ghz but is locked at 2ghz, are you being cheated or getting what you paid for?

If you buy a car with a speed limiter are you being cheated or can you remove it ?

This ruling says you have the right to remove that limit.

I'm not arguing your question. I am calling it a lie to say Tesla was ransoming extended range.

They weren't ransoming anything. They gave the customer the range they paid for and never took anything away.

I said in another comment: ransom is the part you do after you took something. It doesn't describe the taking part.

The context here is more like https://en.wikipedia.org/wiki/Threshold_pledge_system#Ransom...

And my point is that nothing was taken, therefore it cannot be ransomed.

This isn't a Threshold Pledge System either because there's no contracted good being withheld.

> The differentiator the consumer is paying for is range, not # of cells in the battery.

It's not that simple. If your battery breaks out of warranty or starts loosing charge, you can't go to Tesla and say, "hey, I paid for X range and I'm not getting it anymore, fix it." If you do, they'll say, "sorry, physical batteries decay, not our problem", which most of us would agree is a reasonable response. Tesla gave you the hardware, it's not their responsibility to make sure you take care of it.

So in this scenario you're not just purely paying for range, you're also paying for hardware, which has its own warranty period and physical constraints/downsides that Tesla is not responsible for.

In the same vein, if somebody came out with magical tires for the Tesla that managed to effectively increase your range by a substantial amount by having better grip or less friction or something, you wouldn't expect Tesla to go to drivers and say, "hey, you're paying for range. You can't install those tires because they would cause you to get more range than you payed for!" Most of us would call that reaction unreasonable.

So in that scenario, you're also not really paying for the actual range you get, you're paying for a kind of "potential" range that could be increased or decreased depending on other factors outside of both yours and Tesla's control.

The reality is you're not really paying for the range or the # of cells. You're paying for a kind of bizarre combination of both. The physical hardware is your responsibility to take care of, but also you're not allowed to use it in a way that Tesla doesn't like. If you come up with an unrelated way to increase your range, you can do that. You haven't signed a contract that says, "I will stop driving if I've gone more than X miles without a refill, period."

It's understandable to me that different people have different opinions about how those two things should be combined and where it's reasonable for each party's responsibilities to end. For example, the CPU scenario you've listed is fairly common, but most people don't mind because it's not illegal to overclock your CPU. That's a big differentiating factor.

I understand that you disagree with the word "ransom" but it seems pretty obvious to me that that's an opinion, not a statement of fact. The original poster was using a shorthand to suggest that Tesla's attitudes toward consumer rights were unreasonable. I don't see a problem with that.

Technically you're paying for a battery capacity measured in kWh, not range or # of cells.

You do raise an interesting point though about failure. Cells don't fail at the same rate and packs are designed to disable cells (possibly groups of cells) that aren't performing within acceptable operational parameters.

So given that these "ransom packs" are capable of 75 kWh but only deliver the 60 kWh the customer paid for, how do failing cells impact overall capacity?

Since over charging and completely discharging lithium cells shortens their life does that mean because they're never allowed to use their full potential would they actually last longer than a true 60 kWh pack?

> Since over charging and completely discharging lithium cells shortens their life does that mean because they're never allowed to use their full potential would they actually last longer than a true 60 kWh pack?

I am not a mechanic, but off the top of my head, they should last longer, depending on how Tesla has implemented the limit. If Tesla put the artificial charge limits in the middle of the actual capacity, then that would mean you never discharge the last 7 kWh or overcharge the last 7 kWh. If they just make the battery shut down 15 kWh early, then I doubt the gains are as significant, since plugging in your car overnight will still charge it all the way to the top (which from my understanding is the more harmful thing to do).

I'd be very interested to see some research done by someone who knows more about batteries than me, would make a cool technical breakdown. It seems to me that (just like CPU throttling) battery "throttling" could potentially be a useful way to decrease failure. Of course, if you think back to Apple's kerfuffle with throttling iPhones as batteries aged, you can have a good, justifiable customer result that still makes customers angry if they're not informed or suspect you have other motivations.

To bring this conversation back to the original point of the entire thread, it is (for now) legal to circumvent DRM for maintenance. So question: your battery starts failing on your Tesla car -- you buy a completely new battery from a non-Tesla producer that supports 75 kWh. You get a non-Tesla mechanic to install it. As part of the installation process, you break the DRM and allow the battery to be used to its full capacity.

Problem? No problem? When you bought a Tesla, were you paying for a 60 kWh rated battery or were you paying for a 60 kWh car?

Followup question, suppose you find out that Tesla (or any other electric car company with similar policies) has implemented battery throttling poorly for your car. Rather than rotating cells or putting a limit on how full cells get, the battery throttling just disallows a few specific cells from getting power -- so effectively, its as if you literally have a 60 kWh battery with a few (disabled) extra cells stapled on. Your battery range starts decreasing over time, and a mechanic breaks the DRM, which restores most of the failed capacity, which puts you back up to close to your original range.

Problem? Have you violated your contract if the car battery never got above 60 kWh?

And of course, finally you have obvious question, which is, 'if you buy a Tesla car and immediately break the DRM as soon as you get it home to get extra capacity, is that a problem?' Which, yeah, that's a problem. Tesla will not be happy with you.

The difficulty of looking at Tesla's policy as if it's a pure contract is that the physical constraints get in the way -- in other words, it's not a completely encapsulated system. If you look at something like being charged per-mile in a taxi, it's easy to completely divorce that from the physical process. You're charged per-mile that the taxi drove. Doesn't matter how it drove, doesn't matter what the actual cost of driving it is, doesn't matter what the implementation details are -- you're being charged for a result.

Similarly, if you buy something physical, then the physicality provides a reasonable set of consistent restraints and rules. Doesn't matter if it breaks later, doesn't matter why you bought it. In that case, only the physical reality matters and very few contractual things get in the way.

But with Tesla, you've got both systems clashing with each other. It makes sense to say, "okay, you payed for a 60 kWh battery and got a 60 kWh battery." But you didn't really get a 60 kWh battery. You got a 75 kWh battery that has extra restrictions. So you can't completely ignore implementation details, because you're still buying a physical battery, but neither can you take the entire system apart or universally mess with those implementation details.

In the scenario you proposed I think this law would apply, however I don't think that's how it works.

I think the capacity limit isn't a function of the car itself but rather the battery pack installed.

In other words, if I bought an after market 200 kWh battery pack 5 years from now and installed it in my 60 kWh Tesla and it registered as a 60 kWh battery pack, I would burn Tesla to the ground.

Even if I buy a CPU that will melt at 2.5GHz and I can't[0] overclock it to 2.5GHz I'm being cheated. Nominal clock rate is just that: nominal.

0: given sufficient digging around in low-level settings and availablity of a appropriate clock source on the motherboard

Will this ruling now allow Tesla owners to fix this firmware range limitation of their cars?

I don't think so because it's not a fix because nothing is broken. Now if the 75 kWh battery pack degrades to say 59 kWh of capacity and you only get 47 kWh then I think it becomes a question of:

Did Tesla sell you a 75 kWh pack that will only deliver 80% capacity?

Or did Tesla sell you 60 kWh of power?

I would say it's the later and you're with-in your right to circumvent the DRM and boost your failing pack back up to 60 kWh or less.

So, you agree with everything quadrangle said, and that means he's lying?

It's not ransom, it's giving the customer what requested at the price they agreed to pay. And the added bonus of a cheaper than replacing the whole battery upgrade option in the future.

They didn't sell them a 300 mile range then knock it down to 250 when they rolled off the lot. They didn't send them a letter saying "if you want your range back send $$$ by 10pm tomorrow."

The customer paid for 250 miles, they provided a car that could go 250 miles and additionally said "if you would like to upgrade to 300 miles we can do that for less than the cost of a full battery replacement."

So DRM to artificially limit the driving range of their cars?

I'm not a lawyer but I would say so. The question isn't what they did but rather is it ransom or wrong?

Let's clear up a misunderstanding: ransoms aren't necessarily and always wrong. There's indeed some implication that way, but you could just as well argue that this type of ransom is justified.

https://en.wikipedia.org/wiki/Threshold_pledge_system#Ransom...

Is this true? There are mechanics everywhere, so why would every one need to learn to program something that could be made by one and sold as a commercial (or open source) project? Sure, there are many types of cars and situations that need handling, but there are far more mechanics than that.

I'm genuinely asking this because I often question whether it's true that "everyone should learn to program", or it's more like, "we have a programmer shortage".

I think in cases like this they mean "programming" in the same way we use "cooking". Almost everyone can 'cook' at least a few things but only a few people are actual "cooks".

You could buy a commercial product to do what's fairly trivial (write to a couple of HW registers) that only support select models (because they all have their own registers/addresses). You could also just learn what HW registers are and some general techniques to change them and figure it out for any model that rolls in the shop as you need it. The better mechanics can even extrapolate fixes to completely new but related issues.

> I often question whether it's true that "everyone should learn to program", or it's more like, "we have a programmer shortage".

I don't think the parent was saying either of those things. Cars are no longer mechanical devices; they're full of computers now. In order to service any modern car, you need an increasing level of computer tools & training. This has been a growing problem for a long time for independent mechanics who have had no choice but to refer people to the dealerships, due both to proprietary digital systems in the cars, and to the copyright laws that prevented legal home and third-party repairs.

> so why would every one need to learn to program something that could be made by one and sold as a commercial (or open source) project?

Maybe this is a theoretical question, as opposed to a practical one? Car manufacturers today are, by and large, not shipping open source products, and until now it has been illegal to reverse engineer those products, so no open source project could legally exist.

* BTW, total tangent of a side note, but might be interesting to you: the US Department of Defense (and I think other branches of the government too) define "commercial" software in a way that includes open source, in other words open source is commercial. The reason is that OSS comes with a license, it's not to do with whether the software costs money, nor whether the source code is distributed with the software. https://dodcio.defense.gov/open-source-software-faq/#Q:_Is_o...

i started learning python in my spare time after I attended a training course for shop software that ran on Linux. As for the pi, theyre awesome. for $30 or so I have a computer powerful enough to pull OBD codes and do pretty much anything else in the shop wirelessly. And if someone drop it, backs over it, or splashes it with friction modifier or something I can just ask my boss for $40 to replace it instead of $1400 for a new laptop.

Can you point me at some good tutorials on making a pi read ODBC codes?

Did you mean OBD(2)? There are Bluetooth, wifi and usb controllers available that you can use to read and write data. Some of them are very cheap, not sure how well they work though.

https://www.macchina.cc/

In order to change the brakes on the rear of a Volkswagen you need a computer in order to retract the electronically adjusted calipers. This doesn't answer your question but it does exemplify the current situation. Cars aren't purely mechanical devices anymore. Computers play a large role in their operation.

https://en.wikipedia.org/wiki/Electric_park_brake

I find this a bit scary --- mainly because I'm used to cars where the (mechanical) parking brake also can be an emergency brake, and the solid linkage between the lever and the braking mechanism means it is still usable even if all other power is lost. (It's also analog, unlike the "push to set parking brake" on/off behaviour --- especially useful for modulated emergency braking.)

That's certainly true for common problems on common vehicles, but I imagine there is a growing niche of mechanics understand the computer systems behind modern cars more deeply.

Will that commercial product still be sold in 30 years to help repair a "classic" Tesla?

I'm a biolermaker¹ with 20 years experience in the trade.

Now I operate a laser cutter, have a fairly extensive AutoHotKey script, can navigate my way around the post processor that generates the particular dialect of G-Code our laser prefers, and know a touch of python.

I'm definitely an outlier, but I'm not the only computer-nerd-come-metal-fabricator I've worked with.

1. https://en.wikipedia.org/wiki/Boilermaker

* boilermaker, I thought the italics meant it was intentional and was really interested in what you did. I still find it fascinating but wonder what a biolermaker would do?

The Bio/LeR industry is very innovative but also very secretive so I doubt you will find very much about it.

It’s an antiquated term that nowadays just means metal fabricator.

I’ve never made a boiler. I probably could, but I’m not even qualified to weld pressure vessels. I’d pass the weld tests with a couple of days practice though.

I got serious enough about metalwork to start buying machines about three years ago, but became interested in such things helping a friend who bought a used CNC wood router on a bit of a lark. Spent about a month generating gcode in Perl, and then decided to learn how to use Inventor.

And now I have several machines that one generally doesn't find in a second-story urban walkup. (Getting a 400-ish-pound lathe up the steps was how I found my limit there.)

There are very few mechanics who will actually fix parts unless it is for their own vehicle. Most will just quite the price with a replacement part and move on. Even those that know how to do stuff like this rarely will agree to actually repair anything.

I noticed going from my first second hand car probably built in the early 90's or earlier, to my next one from 2000 that there was a lot less stuff that i could fix myself - due to the computerised nature of many things.

I wonder of there would be a market for a purely mechanical (modern) car like the old days.

(One of the beauties of mountain biking I see being lost is the ability to fix maybe 90% of problems with a small pack of tools that you can carry about with you. I can see electric bikes being problematic for that. They will get you further from civilisation and fail worse).

Or just be savvy enough to use what others put out there.

Then again, when I put it that way I begin to worry about mechanics unwittingly infecting cars with malware...

This is very interesting indeed. Please keep talking ;)

If it was legal to do, I'm sure there would be a market so they don't have to, but can readily buy a cheap tool that can do it.

mechanics don't need to learn programming, but there is a growing industry of people making tools for this kind of thing. It has been around a long time in the aftermarket performance world. People have been reprogramming ecus or attaching add on ecus that intercept signals for a long time.

And with hybrid and electric cars, mechanics and auto body shops have to know 3-phase electricity too.

Perhaps you can use a raspberry pi and a bit of python on the current version, but maybe they will be able to improve the "security" on future versions? Having a legal right to bypass DRM doesn't help if nobody knows how.

Also, I wonder how much that $2500 computer and sensor pack costs to manufacture. Perhaps replacing the entire unit with a raspberry pi version would ultimately be the way to go.

Raspberry pies can't deal with sudden power cycles, that corrupts their storage. I don’t think they are suitable for use as a critical part in a car.

i2c.

Congrats, you presented the one and only case where the manufacturers are right: Nobody should tamper with critical control systems.

I don't want a DIY arduino-hacked suspension control chip or ABS. Neither would your clients. Unless of course your client understands this and it goes into the service logbook and you take full legal responsibility for any accidents this might cause in the future.

However I couldn't care less if my mp3 player has it's EMMC swapped or if my laptop has a third-party LCD replacement.

> I don't want a DIY arduino-hacked ...

This is a professional service being done by the same professionals who you're paying to actually replace the suspension, brakes, brake lines, or other life-critical components. If you can't trust them to make a good judgement on how or whether to reset tamper detection, then how can you trust them to replace your brakes or suspension in the first place? Surely the idea is that some company would develop and test a process for resetting the tamper detection now that it is apparently legal, and then this tool would be provided to mechanic shops.

> and you take full legal responsibility for any accidents this might cause in the future.

Do mechanics take full legal responsibility for any accidents caused by replacing your brakes or other things? I'm sure in cases of egregiously negligent behavior, they could probably be sued, but most of the time it would be very difficult to prove that they were the cause. I don't think this would be any different.

> Nobody should tamper with critical control systems.

If you believe this, it sounds like you buy brand-new-only cars the moment your existing car turns on any kind of fault light. But, assuming you're okay with them swapping in a $2500 replacement instead, do you believe in this idea so strongly that you would pay 10% to 20% of the vehicle's MSRP any time a fault triggers? Keeping in mind that a vehicle's MSRP has very little to do with its used market value. $2500 could easily be 30%+ of a vehicle's actual value at the time that this occurs.

I get where you're coming from, but I think the strength of your language is out of proportion with the suggested repairs.

He has a point though... Cars are manufactured in such way that when something fails, the failure is not catastrophic. Car might not be operational, but it is very rare that malfunction causes a crash. With SW I'm not convinced that the same can be accomplished. If you add 3rd party maintainers to the mix, it gets even worse.

I still think the right to repair is worth the risk, just pointing out that it's not so clear cut.

I think your and OPs logic has a fatal assumption: that the quality of work done by a mechanic for a fix is worse than the manufacturer's replacement.

There is a possibility of failure in either one, right? It's people doing the work either way.

The mechanic's professional responsibility is to repair the car. If we're assuming they are too ignorant to do that job, then why go to a third party mechanic in the first place?

When ABS systems fault, they just turn in to non-ABS brakes.

What's the worst that would happen with a malfunctioning active suspension system? It wouldn't lower when going highway speeds, so you'll get more drag.

The only computer that absolutely needs to work to keep your car running is the one that's controlling the EFI system. If that's not running well, it's pretty obvious since the engine will stall out. If you have a modern automatic transmission, then you'll need that one too, and that's pretty obvious if it's not working correctly. Everything else is an accessory, non-essential and part of planned obsolescence.

ABS not working as intended could be a surprise to the driver and cause a crash, especially if you become dependent on it.

It's just like that AF 447 crash, they stalled from 38,000 ft to the ground because they never expected the computer to let them stall the plane.

But I mostly agree, the only software things I would see as absolutely life critical on a car would be drive-by-wire controls. All the rest will just either stall it, reduce performance or break something.

>especially if you become dependent on it.

Nobody is ever going to become dependent on locking up their brakes during normal driving. Power brakes would be dangerous if they failed because without them you get very substantially reduced braking but in no way would the lack of ABS be any more dangerous than driving a car without ABS to begin with.

An ABS failure could easily become fatal if it increased your stopping distance beyond what you'd achieve without it. After all it works by modulating the signal you send to the brakes.

That's irrelevant and it doesn't change the fact that no one drives with the intention to brake hard enough that ABS activates. There's zero chance of anyone becoming dependent on it.

My point is that ABS failure isn't equivalent to not having ABS at all - it can fail in a way that is actively harmful.

I'm not sure that's even possible. I could be mistaken but brakes are designed to be resilient to failure to the point that there aren't really any common single points of failure in a modern car. Brake circuits are actually kept separate from the master cylinder on down so that if one of the seals blows out on the master cylinder there's still another seal for the other circuit. If there's a cut in one of the brake hoses and the fluid leaks out, the brake fluid reservoir splits into two sections at the bottom so that even if one leaks to the point of being bone dry the other side still has some fluid left. The brake booster is designed so that it's a solid piece going all the way through it so if the brake booster is totally broken it works like a linkage. The brake booster is also designed with a check valve so that even after the engine is turned off you still have some vacuum stored in the booster for a couple presses of the pedal before the brake assist stops.

I'm not saying that there couldn't be a failure mode that would prevent someone from using the brakes but due to the attention to detail for everything else about the design and redundancy of the braking system on a car I think it's doubtful that the ABS actuator would be designed in such a way as to allow a software error or broken part to potentially prevent the brakes from working.

There is a lot of snow where I live and the ABS triggers all the time and yes some people just slam the brakes expecting their ABS to optimally slow down on ice.

Yep, and those people are dangerous idiots. Even with ABS it is improper to slam on your brakes on low-traction surfaces. The purpose of ABS is to prevent wheel locking under hard braking, but it is not intended to respond and recover from pure idiocy.

The correct way to apply the brakes is the same with or without ABS, you should always do it smoothly with increasing force as necessary. Doing this would allow ABS to function properly, and in the case of no-ABS, will allow you to feel and respond when locking may occur.

Even if your ABS were to fail, that does not absolve you of the responsibility of driving like a complete moron. It's an unfortunate reality that most of the people on the road in the US frankly should not be driving.

> ABS not working as intended could be a surprise to the driver

That's why there's a warning light for this

In this case the only thing that is changing is a fault state and standard circuit components, not the code running on the device or the device being used. As long as the components are within spec the only problem could be environmental damage from the device being unsealed. I agree with some of the points you made, but not in the context of the previous post.

Reply on news.ycombinator.com