5 ms·
Disclaimer: My father worked as an engineer for John Deere for 35 years. I interned at John Deere writing code that runs on the tractor controllers. Discussion
by bricej13 7y ago
Disclaimer: My father worked as an engineer for John Deere for 35 years. I interned at John Deere writing code that runs on the tractor controllers.
Discussions on this topic always end up one-sided and simplistic. Hopefully I can shed some light on the more nuanced reasoning behind John Deere's position.
Having the DRM in place allows Deere to reduce manufacturing expense and increase platform flexibility. There is a very wide array of needs that farmers have based on what they do. Deere allows buyers to customize tractors to their needs for everything from engine horsepower, to wheel count, size, and type, cab quality-of-life, to hydraulic hookups for implements. Some of these changes are just a software change, while others are a software + hardware change.
Engine horsepower, for example, can be increased by a software update. Techincally, this is pretty cool. Designing and manufacturing engines is expensive. This allows them to manufacture fewer different engines that can cover a wider variety of use cases. It also allows farmers the flexibility to upgrade their engine horsepower at a future date. If I remember correctly each extra 50hp above the base costs ~10k, so the large configurations subsidize the cost of the base configurations.
With that understanding, think of how this can apply to Deere's obligations to the EPA or to warranties.
Years ago, farmers found a hack where they could put a resister in-line between the diesel temperature sensor and the ECU and increase their horsepower. The hack spread like wildfire. This made the engines run in a configuration that had not been tested by Deere or approved by the EPA. Who would the EPA go after if it had caused emissions issues? Should Deere honor the warranty in this case of those who did the hack? How would Deere know if someone did the hack, borked the engine, then removed the resistor?
Liability is the enemy of automation. Deere has added some automation over the years, allowing the tractors to drive straight down the field without intervention, and executing perfect turns at the push of a button. This is functionality that no companies would let end users change. Much like my dad, a tractor is not a cell phone. Installing a custom rom on a cell phone is one thing, updating the autonomous driving of a 10 ton tractor is quite another.
There's got to be some middle ground, but I don't know what it is.
- jacquesm 7y agoAll of the advantages you list could be provided better without DRM from the viewpoint of the farmer, open standards and all that. The three point hitch is what made farm equipment investments a good destination of the money farmers had saved, not DRM.
- rfrey 7y agoThese are all benefits for John Deere. I don't think anybody thinks there is no incentive for JD to do this. Reduced manufacturing expense - Yup, DRM and platform lock-down (machinery is a platform??) increases profits. Tractors sure aren't getting cheaper on account of these "improvements". "Platform flexibility" is nothing more than the ability for JD to lock out capabilities of the machinery that the farmer supposedly bought, and sell it as an add-on later. Again, no doubt this is advantageous for JD, just as locking out portions of a game until later payment is advantageous for EA. It is still terrible for the farmer.
- bricej13 7y ago> These are all benefits for John Deere. Yep, that was my point in posting. It's usually good to hear both sides of things. > Yup, DRM and platform lock-down (machinery is a platform??) increases profits Reducing manufacturing costs is not the same as increasing profits. They're not a monopoly, they still have to set prices according to the market. This allows them to price things lower. > "Platform flexibility" is nothing more than the ability for JD to lock out capabilities of the machinery that the farmer supposedly bought, and sell it as an add-on later. Do you think that when a farmer spends 100k they don't know what they're buying? I don't know what you mean when you say they 'supposedly' bought it. They order a tractor, customize it, and buy it. It allowes them to get a tractor at a lower price point and upgrade as needed
- toufka 7y agoThe market is small. And that means, pragmatically, a professional farmer in search of a combine might be caught between a couple of bad choices - but that doesn't mean they endorse the business practices.
- jwalgenbach 7y agoThe cost to JD to manufacture the tractor with lower horsepower limited by the software lockout is the same as it is to manufacture the tractor without the software governor. That's why it looks like a scam -- JD doesn't spend a dime to deliver the upgrade. They get paid for removing the artificial limits on the product. No one is saying that different hardware (or even software) features of the tractor shouldn't be configurable and charged accordingly, but the lock-out is different. JD is crippling the capabilities.
- driverdan 7y ago> If I remember correctly each extra 50hp above the base costs ~10k Upping the fuel pressure and changing some maps should not cost the consumer $10k. The markup in that is ridiculous. > Should Deere honor the warranty in this case of those who did the hack? So long as the hack didn't cause the problem they are legally obligated to honor the warranty.
- jelliclesfarm 7y agoIndustry standard is $1000 per horsepower. Belarus tractors made in Minsk offer at least 30-40% less markdown after 120 HP. Available in Canada.
- bricej13 7y ago> So long as the hack didn't cause the problem they are legally obligated to honor the warranty. I agree. The problem is that you could update the software to void the warranty, then do a factory reset when you have problems. Deere would never know that you operated the tractor outside of the parameters that they designed, built, and tested, but then they would be on the hook for it. > Upping the fuel pressure and changing some maps should not cost the consumer $10k. The markup in that is ridiculous. My numbers probably aren't that accurate, but they're not selling just software in that case. They've already bought an engine (for cheap) that is capable of that horsepower. They just don't realize the cost for it until they actually upgrade.
- mcclell3 7y agoIt likely costs 10k at an average. The point isn’t that it costs $0 to upgrade software. It is that Deere gives a lower price on high spec bc they under price the low spec. They get higher volumes on engine technology bc they use software to meet many customer’s power needs. If Deere made made a perfectly sized engine for every make and model with no software variation they would be idiots and have even higher priced machines. I have no love for the monopoly these companies have on the farmers of the world but the people against Deere here are drastically simplifying agriculture’s OEM business model. The simple part of this argument is the right to repair part. Deere should just sell their service tool at a profit and provide software updates to the customer and not just dealers. Then we would be done with this right to repair argument.
- jelliclesfarm 7y agoYes. All of this favors JD.
- towndrunk 7y ago"Having the DRM in place allows Deere to reduce manufacturing expense and increase platform flexibility" Sounds like it's right out of the sales brochure.
- count 7y agoWhy are the tractors any different from cars? If I buy a car, I can do anything aftermarket to it I want. If the car is in warranty, those mods do not allow the manufacturer to not continue to support it (well established case law), and the EPA doesn't care once it's been sold (individual states like CA might have issue, but those are with the owner, not the manufacturer, if it's been modified). This is all well-settled in the automotive world, why is JD 'different'?
- hrktb 7y agoCar manufacters is an interesting case. Here in the EU Volkswagen has a different maintenance and repair policy depending on whether the car stayed in the network for revisions or not, and I think aftermarket changes would also matter. They’re not refusing or blocking anything, just putting a price tag on consumer’s freedom to do what they want.
- naikrovek 7y ago> This is all well-settled in the automotive world, why is JD 'different'? Because laws were passed in the US that specifically allow these kinds of things (like customer repairability and modification) for cars. Those laws do not apply to tractors.
- magduf 7y agoYes, actually, they do. The Magnusson-Moss Warranty Act of 1975 applies to everything, not just cars, even though it was really aimed at cars. But it applies to anything consumers might buy, whether it's tractors or TVs.
- war1025 7y agoYou are much more likely to call dealer/tech support for a tractor than you are for a car. It's the same reason enterprise software contracts are mainly just paying for support. If your business is on the line, you want it to work, and you want someone to fix it when it's not working. Tractors are basically automated plant-production factories. They have tons of sensors and moving parts. If you want support from the people that made it, it's not unreasonable that they can dictate the terms of use in order to get that support.
- x86_64Ubuntu 7y agoWhy does there have to be a middle ground? Things seemed fine before all the DRM. And I'm not sure too many of your things actually REQUIRE DRM to be carried out. And as far as your warranty FUD, it's known in other fields that if you make an unapproved modification to an item, or have it done by someone else that it voids the warranty. This is a problem that's already been solved that you are trying to use to support the case for DRM.
- hilbert42 7y agoDRM? It's simple, enforcing DRM means more money to JD. Even those with a room-temperature IQ know that!
- bricej13 7y agoI must have done a poor job at communicating if you think I was trying to support the case for DRM. My purpose was to help people understand where Deere is coming from and some of the details that the "John Deere Bad!" comments just don't capture. > And I'm not sure too many of your things actually REQUIRE DRM to be carried out. I actually agree. They're using DRM as a legal fix for what is a technical problem. I feel like you missed the point about the warranty. We probably agree that if someone re-flashes their controllers they've voided their warranty. The problem for Deere comes when someone screws it up, flashes back to factory settings and takes it in for warranty work. They could have done tens of thousands of dollars of damage. This is not FUD to Deere, it is an actual risk. Their 'fix' for that problem is DRM. (Again, I'm not promoting DRM)
- ddingus 7y agoSure seems to me Deere could implement an incremental state storage of some kind to get around the "flash it before fix it under warranty / annual maintenance" problem. It also seems to me Deere could simply quote the repair too. Hell, bundle it in with financing so the user learns their lesson without breaking the bank. So, farmer bob mods his machine, breaks it. I know Deere does a ton of actual hard testing. They know what, "damn, where did that log or boulder come from" looks like. They also know what, "it is running hot / over spec" looks like too. Deere sees a freshly flashed machine, and it goes as follows: Why did you flash it? Why did you not call us first so we understand what happened? That conversation ends up either a warranty / maintenance conversation, or not. Then, the work to put the machine back to spec gets quoted. Farmer Bob pays up, perhaps with a loan, and everyone moves on.
- TeMPOraL 7y agoThanks for posting this. I'm generally biased against DRM as it tends to enable abusive business practices, but you've made some solid points I wish people engaged with better in the discussion. The liability angle is particularly interesting here - I didn't really consider that part of the drive to DRM everything might be pressure created by regulators. It's obvious in the case of autonomous driving, but not necessarily elsewhere. Then again: > Who would the EPA go after if it had caused emissions issues? Would they really go after Deere? I never checked this, but I think in case of cars, end-users are liable for the modifications; if the company has tests proving that the model under investigation meets the regulatory standards in its sold configuration, then they're off the hook. Why would the same thinking not apply to farm equipment? -- In the perfect world, it would all work itself out. In the real world, while the farmers have every incentive to extract as much performance as physically possible from their equipment, JD - like every business - has a lot of incentive to screw farmers over. Competitive pressure is a traditional protection against too much abuse of customers, but it doesn't really apply all that much when you have a small amount of providers. DRM itself is, in its general form, a mechanism for creating a localized alternative reality, in which you can attach colour to bits[0]. It allows businesses to enforce arbitrary rules in their products - rules that would be impossible to enforce in pre-computer reality. This is open for abuse, and also kind of destroys the protection of competitive pressure - attach DRM to something, and its complementary commodities stop being commodities. Customers lose their traditional protection from abusive tendencies of for-profit businesses. There must be a better way, one that creates a fair balance between interests of sellers and buyers, but I too don't know what it is. -- [0] - https://ansuz.sooke.bc.ca/entry/23 https://ansuz.sooke.bc.ca/entry/23
- bricej13 7y ago> Would they really go after Deere? I don't know the details, just that this something they worry about. I'm guessing that <speculation> the wording in the EPA regulations requires companies to take 'reasonable measures' to prevent end-users from subverting emissions mechanisms. 'Reasonable measures' is a grey area that the lawers get to fight about. That said, buying a black box off of ebay and plugging it in seems pretty easy.</speculation> > In the perfect world, it would all work itself out. In the real world, while the farmers have every incentive to extract as much performance as physically possible from their equipment, JD - like every business - has a lot of incentive to screw farmers over. Competitive pressure is a traditional protection against too much abuse of customers, but it doesn't really apply all that much when you have a small amount of providers. DRM itself is, in its general form, a mechanism for creating a localized alternative reality, in which you can attach colour to bits[0]. It allows businesses to enforce arbitrary rules in their products - rules that would be impossible to enforce in pre-computer reality. This is open for abuse, and also kind of destroys the protection of competitive pressure - attach DRM to something, and its complementary commodities stop being commodities. Customers lose their traditional protection from abusive tendencies of for-profit businesses. Well said
- zelon88 7y ago> Engine horsepower, for example, can be increased by a software update. How do you see this as an asset? Deere designed a part that's capable of doing something and your software chokes that back unless they poney up. In manufacturing we have the same thing. You buy a CNC machine that comes equipped with 16mb of memory, but only 2mb is usable. If you want all 16mb (which already exists soldered to your motherboard) you need to pony up thousands of dollars for a 16 digit code that unlocks the added memory. And you're trying to tell me that by requiring the manufacturer to share that code is bad for the consumer? Yeah, you don't sound like a shill or anything. And no, Deere shouldn't fix that under warranty. It's the same with cars. You can do that trick to a Honda with a resistor in-line to the MAF sensor and it will run the car lean, giving the illusion of more performance while wearing out the engine and burning the combustion chamber way too hot. Should Honda fix that? No way! Should Honda let the customer do it anyway? Of course they should! Should Honda share the schematics with the customer so they not only realize that it's a bad idea, but also know WHY it's a bad idea? Yes. Your argument is straw man. If you didn't have secrets you wouldn't need DRM. DRM doesn't protect anyone except the edge-case of idiots who shouldn't mess with the tractor even if you gave them the repair manual anyway. It's strictly to protect Deere.
- bricej13 7y ago> your software I was an intern. I wasn't anywhere near the engine controller software. > And no, Deere shouldn't fix that under warranty. It's the same with cars. You can do that trick to a Honda with a resistor in-line to the MAF sensor and it will run the car lean, giving the illusion of more performance while wearing out the engine and burning the combustion chamber way too hot. Should Honda fix that? No way! Should Honda let the customer do it anyway? Of course they should! Should Honda share the schematics with the customer so they not only realize that it's a bad idea, but also know WHY it's a bad idea? Yes. Looks like we agree here. > It's strictly to protect Deere. We also agree here. > Engine horsepower, for example, can be increased by a software update. > How do you see this as an asset? I explained this further in another comment. > Yeah, you don't sound like a shill or anything. I explicitly stated my bias in the first line.
- 7y ago
- cellularmitosis 7y ago> Years ago, farmers found a hack where they could put a resister in-line between the diesel temperature sensor and the ECU and increase their horsepower. The hack spread like wildfire. And no wonder. Inducing artificial market segmentation by selling software-crippled devices is hugely unpopular with consumers. I find it ironic that the unlock ended up being a resistor, just like with the Promise IDE controller / RAID controller from long ago -- another example of a company trying to save costs by designing one product and shipping it as two products (with one of them software-crippled). That instance was also hugely unpopular with consumers, and that resistor hack also spread like wildfire.
- war1025 7y agoBased on the way the comment was worded, it wasn't a power upgrade that was turned on/off based on a resistor being present or not. Adding the resistor changed how the engine thought it was running, which caused it to run in a way that produced more power, but not in a configuration that had been tested or approved for production use. There are lots of things that you can run outside of spec, but they introduce premature wear, which is why they don't run that way by default.