so much this. I don't want to be forced into sacrificing what's possible to keep myself safe, thank you very much. I'm more than capable of performing my own risk assessments, and the further we go down the "secure the hell out of everything" road the closer we get to industrial capture of essential tech.

I think it was a piece by stallman I read recently, about the concept of hardware signature-key application whitelisting. kept me up for a week.

Lay off implications about bad faith, please.

The argument is not that general purpose computing was a mistake, but that general purpose computing is not necessary and is to some extent counterproductive for the majority of consumers. Consumers moved away from flip phones because they wanted more capable phones, that doesn't mean they largely want capability to the extent of general purpose computing.

Consumers will be the judge of whether a platform or product is worth using. Given the unparalleled success of iOS and Apple's locked down ecosystem, it's pretty clear that many find this level of security is very much worth using regardless of imperfect moderation.

> Lay off implications about bad faith

You're right, I crossed a line there.

From a market perspective, the problem is that in the short term it might be feasible to build a closed, tightly controlled market that rivals open alternatives, but in the long term general purpose computing acts as a safeguard against market capture and anti-consumer behavior -- and to a certain extent, consumers and markets in general are very bad at optimizing for long-term consequences.

The movement of both iOS and Android in this direction would not be as concerning if they didn't hold a duopoly over the entire smartphone market. Apple in particular has faced significant antitrust criticism in this area.

Open markets that empower consumers (and I mean that generally -- not just general computing but also home cooking, self-publishing, self-repair and hardware DIY) are the reason why closed-down markets don't degrade and become awful over time. Almost every capability on the modern locked-down iPhone started out as a 3rd-party proof of concept that Apple was later forced to offer in-house alternatives to in order to remain competitive.

It benefits even normal users who don't care about general-purpose computing that there be at least one mainstream option on the market where users can fix their own problems without asking a company for permission. And I don't think it's an accident that as Android and iOS have both moved away from that role, that we are now seeing increased calls for antitrust, increased criticisms from developers, and general outright rejection from these companies of new innovations like game streaming.

> Given the unparalleled success of iOS and Apple's locked down ecosystem, it's pretty clear that many find this level of security is very much worth using regardless of imperfect moderation

I do think it's slightly problematic to assume that users are conscious enough of security to make an educated decision to opt into a locked-down platform, but are not educated enough to avoid flipping a switch in the settings that turns that environment off and on. I don't think that users temporarily become security conscious only when they're in the act of purchasing a phone.

The more likely reality is that most users don't think about security or 'openness' at all beyond reputation/advertising, and the vast majority of iPhone users have probably never thought about the tradeoff between open access and security at any point during any of their purchase decisions for any computing device.

> General purpose computing is one possible safeguard, it also has pretty big and clear downsides for many consumers and isn't obviously the right choice for most.

Yes, but isn't it sad if (say) mom and dad have better hardware at their disposal than people who need general purpose computing for their jobs (e.g. scientists, hackers, ...).

Optimizing for the majority is not always a good thing as it can result in bad outcomes for minorities.

I think it would be sadder if mom and dad had terrible security and lost their savings or treasured memories as a result. Optimizing for one group is always going to cause some problems for a different group, and if I had to choose I'd optimize for the majority every time.

But you're sailing along with a false dichotomy. We can have hardware that is safe for mom and dad, and open for professionals.

I'll just have to disagree that it's a false dichotomy. System openness is diametrically opposed to security; the more open a system is, the more insecure it becomes. This is especially true for mainstream consumers who have little knowledge or skill to administrate open systems securely. It's one of the classic engineering trade-offs, thinking you can have your cake and eat it to seems pretty ignorant to me.

> the market will eventually reflect and account for that in the future.

Which won't help unless we're willing to break apart duopolies and enforce government antitrust. The point I'm making is that when you get rid of consumers' ability to solve their own problems, they lose the ability to solve their own problems.

They don't magically get that ability back when the market starts being terrible. Take a look at Amazon's DRM -- it doesn't matter if you as a consumer wake up one day and realize that there are negative consequences to being unable to port your library to any other devices. You still can't do it.

Market capture is not a problem that can be solved by the market on its own, which we've seen repeatedly throughout the history of US markets, including in the computing market.

This is why we have regulations around some of the most egregious anti-free-market activities companies can take. For example, it's illegal to use warranties to block unrelated consumer repairs. Car makers are legally required to use some universal interfaces that allow non-manufacturers to repair and alter the vehicle. And we're currently campaigning to get rid of DMCA restrictions on breaking DRM for legal purposes like porting Kindle books to other platforms. None of that is stuff that consumers on their own would prioritize, but they're market conditions that benefit everyone tremendously. These are instances where the free market can't solve the problem on its own, there have to be legislative changes that allow the market to compete.

And unless you're currently buying stock in Purism or Windows Phone, I think we both know that the current smartphone market is not set up to allow competition.

> which users happily and quickly disabled to allow malicious programs that they didn't really evaluate at all

So what makes you think they didn't? You're assuming that consumers are making a rational choice when they purchase a phone, but not when they use the phone. I don't think people's brains stop working when they turn their devices on, I think that we should apply a consistent framework to understand both people's computing usage and their purchasing decisions.

> Knowing when to disable security guardrails, however, requires actual security knowledge

No, it really doesn't. You can have a big warning that says "this makes your phone insecure" and people don't need to know the details to trust you.

Of course, in practice, people ignore those warnings. But there's two ways to interpret that -- either people don't understand security/access at all and we shouldn't treat any of their purchasing decisions on this with reverence, or people are making a security decision not to trust phone manufactures when they uncheck that box, and we shouldn't shame them for having a different risk model than us.

I object to any attempt to try and characterize them as somehow being both conscious/unconscious of the risks at the same time, there has to be some consistency in how we talk about those people. How do you know that normal users don't just have a separate threat model than you and that they're willing to uncheck that box because they're consciously deciding to tolerate a greater rate of infection/malware than you find acceptable?

> And unless you're currently buying stock in Purism or Windows Phone

OT but if Microsoft released a reasonably-open Linux phone, all would be forgiven as far as I'm concerned. They've always made fantastic hardware, and they're one of the few companies who could potentially break into the market in a real way.

> Take a look at Amazon's DRM -- it doesn't matter if you as a consumer wake up one day and realize that there are negative consequences to being unable to port your library to any other devices. You still can't do it.

Because almost no one actually cares that you can't port your library to any other devices, Amazon's DRM is well known by now and they certainly don't mislead customers into thinking their books are DRM-free nor do they advertise library portability as a feature. The mechanism by which consumers "get that ability back" is by buying DRM free books if they're sick of DRM, but that hasn't and doesn't happen for most consumers even though they're completely free to do so.

Most people are content with Amazon DRM, and when they aren't, any company that sells DRM-free books will make a killing. But no one makes a lot of money from DRM-free books because there is no widespread demand for it. It's ridiculous to suggest that consumers have been robbed of their ability to influence the market, they have always held all the cards. Companies only make money if consumers want what they make, and boy do they want it.

If you have reason to believe Amazon is somehow preventing consumers from buying DRM-free books, I'd be happy to hear it. The kind of failures that markets can't solve themselves are typically ones of deception and coercion, show me that.

> None of that is stuff that consumers on their own would prioritize, but they're market conditions that benefit everyone tremendously.

Says you. I'll let consumers decide what's important and what market conditions are beneficial to them, thank you.

> either people don't understand security/access at all and we shouldn't treat any of their purchasing decisions on this with reverence, or people are making a security decision not to trust phone manufactures when they uncheck that box, and we shouldn't shame them for having a different risk model than us.

People don't need to understand security/access at all for their purchasing decisions to be legitimate. Do you need to be an expert in plumbing to know that you need a plumber? Do you need plumber skills to figure out that the wet & broken mess you've made by yourself suggests you need expert help? You can't conflate a consumer's knowledge of security risk with a consumer's knowledge of security itself. Security risks & benefits are really obvious: consumers notice when they or people they know get hacked & phished. They notice when their computer slows down to a crawl and starts injecting garbage into their screen. They especially notice when iPhones & Chromebooks have none of these problems.

Knowing if a non-vetted program is safe to bypass regular security checkpoints? Consumers consistently and repeatedly fail that test almost every time they're given the opportunity.

> How do you know that normal users don't just have a separate threat model than you and that they're willing to uncheck that box because they're consciously deciding to tolerate a greater rate of infection/malware than you find acceptable?

I know because they switch to a more secure & locked down platform wherever they can. I know because they loudly complain that Android is full of malware and unvetted trash and then proceed to buy an iPhone. I know because the anger& confusion they lash out with when they fuck up their own security clearly indicate that they'd rather never deal with that ever again. The idea that most consumers were happy with the previously awful state of security is totally absurd when they fled those platforms en masse the first chance they could get.

> Most people are content with Amazon DRM, and when they aren't, any company that sells DRM-free books will make a killing.

> If you have reason to believe Amazon is somehow preventing consumers from buying DRM-free books, I'd be happy to hear it.

Look, we can get into the nitty gritty details, but if we're going to have this debate, you have to do some basic research online to see how Amazon is currently hurting the Ebook market, or how DRM is actually affecting markets and consumer choice at this moment. I will still assume you're coming into this from an honest perspective of inquiry, but you have to put in a little bit of work here, I'm not going to summarize the entire history of US antitrust for you.

It's going to be very difficult to have a productive conversation if you don't understand how device/vendor lock-in works. Do you really not understand how holding people's purchased libraries captive prevents those people from moving to new devices or ecosystems? Do you really, honestly believe that a superior book ecosystem of any kind could spring up and Kindle owners would just say, "well fine, I'll throw away $200 of purchased Ebooks! It's no big deal." You don't think it affects consumer's book purchasing decisions at all that it's impossible to put Kobo books onto a Kindle or Amazon books onto a Kobo device without breaking a federal law?

Please do some basic research about how DRM and platform lock-in works on Apple/Amazon devices. I heavily recommend reading some of Cory Doctorow's work, or looking at the history of how the DMCA/CFAA has been used to shut down market competitors via legal means instead of via competition, or into how Apple increases lock-in with their current product direction on things like SSO/subscriptions, credit card offerings, and proprietary device standards that competing companies are not able to interop with.

Apple's current product strategy is to increasingly make it so that anyone who wants to buy a different phone must also throw away a substantial amount of sunk cost into app purchases, hardware that isn't allowed to work as well on other devices, social status (incoming Android messages are literally labeled differently in iOS), they may even be forced to cancel their current credit card. This is a really textbook example of how vendor lock-in works, you can find more detailed information online if it's something that you're actually curious about.

----

The way I see it, there are two fundamental problems I have with your point of view.

A) I don't believe that customers magically become stupid or smart based on whether they're holding a smartphone or a credit card. I believe that people's decision making processes when using devices and purchasing devices are the same.

B) I believe that if people actually want to do something, you (usually) shouldn't have to force them to do it. If people genuinely wanted to be in Apple's walled garden, Apple wouldn't need to lock the garden from the inside. It's one thing to argue that people are too stupid to make their own security decisions. It's one thing to argue that people are too stupid to prioritize long-term market effects. But it's ridiculous to argue that people want to be in a walled garden when literally every single opportunity you give them to turn it off or install apps from outside of it, they immediately do so.

That is not the behavior of people who are happy with where they are. I don't know how to do my own plumbing. The reason you can tell that I'm happy to let other people make my plumbing decisions is because I haven't yet taken a hacksaw to my pipes.

If every single opportunity to take apart my toilet I did so, you might reasonably conclude that I did want to do my own plumbing, however misguided my beliefs about my own abilities were. And if having a checkbox in settings labeled, "warning, you are definitely going to be infected by malware" doesn't prevent people from turning off security safeguards, you might reasonably conclude that they do want to manage their own security, however misguided their beliefs about their own abilities are.

----

> I know because they switch to a more secure & locked down platform wherever they can.

This is a wildly simplistic model of how customer behavior actually works. There is a nontrivial percentage of Apple customers who switch to iOS because they're worried about their SMS message bubbles getting colored green. You can't seriously argue that those people are making a security decision.

And in any case, this is all kind of moot because if you want to pretend the market perfectly reflects majority preferences, then people didn't flee Android en-mass. Apple holds a market majority on paid app downloads, but they don't hold a majority of market share on actual devices. The majority of users switched to Android.

Of course, they probably switched to Android in no small part because of large price differences, not because of "openness", but if I'm correctly understanding your worldview it seems like we're not allowed to acknowledge things like that, we have to assume it was an ideological decision because the majority of the world wanted sideloading.

There's no misunderstanding that DRM prevents ecosystem portability, but consumers know about this and consistently buy into them. Just like how they know that Apple iMessage only works with other iPhone users to the chagrin of their Android friends, and still buy into it, sometimes because of the exclusivity. No one registers for an Apple Card thinking it's compatible with Android. "Superior" DRM-free book ecosystems already exist and you've probably seen people promoting these ecosystems to regular consumers with clear explanations of how DRM-free helps ecosystem portability; If you've ever tried to promote this kind of thing you know the kind of blank stare or apathetic silence this is met with.

DRM is an old technique that's been a staple of digital markets almost as long as digital markets have existed. Switching costs and proprietary lock-in are even older concepts. Every market participant knows about them and invests accordingly. You think when the iPhone came out, people didn't leave anything behind? How did RIM's BBM lock-in work out for them? How did iOS ever thrive without Office & Windows integration, lock-ins that Microsoft tried to exploit? Or what about Internet Explorer, which despite huge antitrust action, Microsoft was allowed to continue bundling and defaulting? How did Microsoft's thought-to-be unassailable lock-in work out for them in the server market? Did Linus go to Congress complaining "how can you expect Windows users to say: well fine, I'll just throw away my Windows software, It's no big deal."? This is what happens when consumers in the server market actually give a shit about openness: a completely open ecosystem actually wins.

There's no lack of understanding, consumers see these walled gardens and say "I want in". They know the switching costs and go in anyway. If low-cost switching was actually important to them they would invest accordingly as consumers do in markets where interoperability and openness is actually important. If their locked-in product is actually serving them poorly, they will actually switch.

---

A) I also don't believe that customers magically become stupid or smart based on whether they're holding a smartphone or a credit card. Here's what I also believe: You can be smart yet still be terrible and unskilled at device security. Conflating the intelligence of a consumer with their security skill doesn't make sense at all. Security is a complex and technical field.

B) So who forced or deceived consumers into buying an iPhone? Who forced or deceived them into buy DRM'd books from Amazon? If people genuinely didn't want to be in Apple's walled garden, they wouldn't have gone in to begin with. If they actually cared about openness, they certainly wouldn't repeatedly buy an iPhone.

No, people don't take every opportunity to install apps from outside or disable restrictions. Jailbreaking was well-known and remarkably easy at one point (I think it's harder now?), and it was popular within certain circles but it was never even close to a majority of iOS users. The number of people who install stuff like F-Droid and sideload apps is a tiny fraction of Android users. I doubt the majority of Android users have ever sideloaded anything. Desktop platforms like Windows/macOS/Linux are much more sideload-happy, but desktop marketshare is losing ground to more locked down platforms like Android & iOS. That's the behavior of consumers who are happy with locked down platforms and don't care for general purpose computing.

> There is a nontrivial percentage of Apple customers who switch to iOS because they're worried about their SMS message bubbles getting colored green. You can't seriously argue that those people are making a security decision.

Yes I can, their decision is "SMS message bubbles are the most important difference to me". For them, other factors like security or lock-in are less/not important. If I willingly choose to buy a product that lacks something compared to competitors, it's a pretty clear indication that is something I prioritize less. That's going to be the case for many consumers, security is not their top concern, but it is usually a factor, especially if there are serious discrepancies between competitors. For example, app selection is usually not a factor because both platforms broadly share major apps and there is little noticeable difference between them. With Android taking steps to move closer to iOS' security model like mentioned in the article, Google is reducing the noticeable difference in security between both platforms so it becomes less of an important factor. Security was far more of a factor in Android's early days when it wasn't rare to hear people note Android's terrible security before buying an iPhone.

In the US, Apple doesn't necessarily hold a large majority but they are something like half the market all by themselves. They are also more profitable, which means consumers value and want them more. It's true that price is one of the largest factors between Android & Apple, but at the high-end range of phones where price is less of a confounding variable, Apple is still the brand of choice. Android has recently moved closer to iOS in security model as mentioned in the article, and I think that helps Android stand stronger against the iPhone at the high-end.

> which despite huge antitrust action, Microsoft was allowed to continue bundling and defaulting?

> how did iOS ever thrive without Office & Windows integration, lock-ins that Microsoft tried to exploit?

You need to go back and reread the history here.

I think you have a fundamentally flawed view of how the market works. I'm not sure what else I can say except that markets don't work the way you think they do.

That's not a Libertarian or Socialist or Ancap thing, you can belong to any political spectrum while still understanding that profit is not a perfect representation of value, and that the market is a system that reflects the education of its participants, and that markets can have unoptimal outcomes sometimes.

In a weird way, what you are saying would be correct if the market worked the way you think it did. Except it doesn't, and you're misunderstanding the history of all of the cases you site (and conveniently leaving out really important stories like Bell Labs and the history of the Internet), and you're creating an alternative history where lock-in just kind of doesn't have any effect on anything, and where the market just kind of magically solves all of the problems it creates, and where we haven't had significant legislation to fix these problems.

At some point you have to go back and critically examine those assumptions, they're just not true. Even Conservative circles are going to laugh you out of conversations if you come in with an economic model that isn't sophisticated enough to explain phenomenons like forced arbitration or externalities like pollution. Perfectly efficient markets do not exist, every market has some inefficiency and some instances of consumers exercising their purchasing power against their own interest.

> They are also more profitable, which means consumers value and want them more.

Again, this is just not how markets work or what profit represents. There isn't a perfect one-to-one link between profit margins and consumer value.

> It's true that price is one of the largest factors between Android & Apple, but at the high-end range of phones where price is less of a confounding variable, Apple is still the brand of choice.

And again, this is not really a valid way to think about market data. Anyway, it doesn't matter because according to your worldview the only thing that matters is what consumers choose, and consumers have clearly overwhelmingly chosen that they want the cheaper phones instead of the flagships.

> For them, other factors like security or lock-in are less/not important.

Right, they're not thinking about security and they don't care about it, they care about the green bubble. The security of the phone is irrelevant, and it doesn't make sense to look at their purchasing decisions as some kind of validation of Apple's security model because they don't care, it is not a factor in their purchasing decisions.

If Apple allowed sideloading tomorrow, their purchasing decisions would not change. So why on earth would we use them as examples to say that sideloading is a problem?

> Conflating the intelligence of a consumer with their security skill doesn't make sense at all.

I don't need to be a plumbing expert to know not to take apart my toilet. Nobody needs to be a security expert to know not to flip a switch or mess with default security settings. It's not a complicated system, when I don't know how something works, I just don't mess with it.

I'm not going to let you walk away from the fact that you're saying customers are perfectly informed about the tradeoffs of buying into an ecosystem, but too stupid to be taught not to press a single, clearly market button buried in a phone's settings. That is a contradiction.

> I doubt the majority of Android users have ever sideloaded anything.

Then why is it a security risk to give them the option? You're arguing that iOS is more secure because it doesn't offer people the option to do something that you don't think they would do.

I don't believe perfectly efficient markets exist either and I'm aware of phenomenons like forced arbitration, negative externalities, and consumers voting against their own interests. But you can't start from "the market is imperfect" and then jump to the conclusion that the smartphone market is broken to the point that it requires government intervention. Honestly, it's a little difficult to discuss this when you just vaguely dismiss everything as "you're misunderstanding the history of all the cases you cite". Elaborating on even one example keeps this grounded, although I understand if you don't think it's worth the effort.

I'll take a crack though: the inefficiency/externalities currently present in the smartphone market are well within the capabilities of the market to correct itself. When I say that consumers are getting what they want, I don't mean "the market is literally perfect and there's absolutely nothing wrong with it". Antitrust & government intervention isn't warranted on every market imperfection and risks distorting the efficiency of the market even further because government action is rarely restrained or nuanced. The tech industry traditionally has relatively little intervention and it is littered with similar examples where market obstructions like lock-in or DRM are eventually accounted & corrected for. Yes, there was that big breakup of Ma Bell which held a unique 100-year monopoly, but I don't think any Big Tech is quite comparable to Ma Bell.

And frankly, consumers voting against their interests is a slippery accusation that should require a high burden of proof. Corporate deception can cause this, but beyond that it's your word against the consumer's. I'm not inclined to believe suggestions that consumers would care about X if only they were better informed. Prove it in the market, one could pass a small law that improves labelling or something and see if consumer behavior changes.

> And again, this is not really a valid way to think about market data. Anyway, it doesn't matter because according to your worldview the only thing that matters is what consumers choose, and consumers have clearly overwhelmingly chosen that they want the cheaper phones instead of the flagships.

A little more detail than "you're wrong" would be appreciated. Yes, consumers have overwhelmingly chosen cheaper phones over flagships, but there are multiple groups of consumers with different choices, not all of them are price sensitive. One thing most groups of consumers have in common though is that they don't care much for openness. This is reflected in low legitimate usage of Android's open features, steady bleed from open desktop platforms, and negligible interest in open mobile platforms (including Android-compatible forks).

> The security of the phone is irrelevant, and it doesn't make sense to look at their purchasing decisions as some kind of validation of Apple's security model because they don't care, it is not a factor in their purchasing decisions.

The security of the phone is not irrelevant, I don't know why you think that. There does probably exist a group of consumers for whom the green bubble is the only thing that matters, but that's not everyone or likely even a majority.

If Apple allowed sideloading tomorrow, their purchasing decisions wouldn't change because there is no non-sideloading alternative. Apple is basically the only brand that has such a tightly controlled walled garden. That doesn't mean they don't want it, and forcing Apple to support sideloading takes away a consumer's right to choose to enter a walled garden like that.

> I don't need to be a plumbing expert to know not to take apart my toilet. Nobody needs to be a security expert to know not to flip a switch or mess with default security settings. It's not a complicated system, when I don't know how something works, I just don't mess with it. I'm not going to let you walk away from the fact that you're saying customers are perfectly informed about the tradeoffs of buying into an ecosystem, but too stupid to be taught not to press a single, clearly market button buried in a phone's settings. That is a contradiction.

It's simply wrong to imply that you have to be stupid in order to ignorantly flip off a security switch. The implications of flipping off a security setting is not obvious to laymen and malicious sites will attempt to provide a plausible explanation for why the switch is okay to turn off, an explanation that you need some technical knowledge to debunk. Fortunately, there aren't many motivated liars that are trying to convince you to take plumbing into your own unskilled hands.

Some platforms end up hiding these options deeper & deeper but it's a Sisyphean strategy. Apple has simply taken the option away entirely it which is the least open but most reliable solution.

Teaching them security is also not only incredibly expensive & difficult but also something most normal people don't really want to learn. Shoving elaborate dialogs and explanations in front of them is an effective way to annoy your users (assuming users don't just skip straight past them), or even worse: a skim read that gives your users a false sense of mastery.

> Then why is it a security risk to give them the option? You're arguing that iOS is more secure because it doesn't offer people the option to do something that you don't think they would do.

The majority of Android users not sideloading doesn't mean that it's not a significant security risk, I don't follow this line of reasoning. A majority of users never get seriously breached, even on insecure platforms like Windows, the remaining group of people that did is still quite significant, certainly enough for it to be a serious problem. It would be an unmitigated disaster if a majority of users were at high risk of breaches. An example I remember: it didn't take that many Fortnite players to sideload & spread malware for people to take notice.