Comment Re: Who writes nonsense like this? (Score 2) 30
RISC-V is also a an arch build target, and it's perfectly reasonable to say that a target platform in a build system supports such and such
RISC-V is also a an arch build target, and it's perfectly reasonable to say that a target platform in a build system supports such and such
At first read, I thought that like in XMPP, some metadata wasn't e2e encrypted (such as the "TO" field). But no: this appears to essentially be a traffic analysis timing attack...
The adversary (govt) has complete control over the network, and sees a message going from endpoint A over the network to Whatsapp servers. Then sees a message of similar (same?) size going from WhatsApp servers to a random endpoint B, and infer that A and B communicate together.
I'm not really sure you can mitigate that without massive consequences for the users (delays or increased decoy traffic). A VPN could get one out of the monitored network, but that's a giant leap of faith.
Step 1: Pad all small packets to the 99th percentile packet size. If the majority of your traffic is the same size or a small number of sizes, an adversary can't use the size of the chunks to reduce the comparison space meaningfully.
Step 2: Send out one packet to the client once per second, no matter what.
Step 3: Defer packets sent out to the client until the next one-second boundary.
You have now completely thwarted timing analysis. Do this by default for all connections. No one will notice the average 0.5 second difference. Want to make things slightly more instant? Change the delay to a smaller number. But really, one second is probably finer granularity than will ever matter in practice, so why bother?
Now that scheme breaks down when images are involved. But that can at least partially be solved by storing the encrypted image on the server until the user takes an action to fetch it, which adds a certain amount of randomness. And pad images to a fixed size to further thwart timing analysis. If people send large numbers of images to specific people, the pattern will still eventually emerge by the timing of people downloading multiple chunks of data, but it will at least be somewhat harder to spot the pattern and will require a larger number of messages.
And if you further distribute the stored, encrypted images across servers that are used for other things (e.g. the servers used for Facebook ads), you can probably make it considerably harder to find patterns unless the government is monitoring two specific people to see if they are communicating.
51% of EVs in the U.S. are Teslas, with ridiculous sensor packages. If it were possible to not hit the pedestrian these vehicles would do so. Yet instead of being safer than ICE vehicles they are 3 times more likely to hit pedestrians in urban settings. The reality is, that according to the numbers, big vehicles are significantly less likely to hit pedestrians. No one is stepping out in front of an enormous diesel pickup truck. But they are stepping out in front of tiny electric vehicles, and they are doing so in ways that even when the vehicle has the best driving assist and sensors on the planet.
The conclusion must either be that people that own electric vehicles are psychopaths (which may well be true), or quiet vehicles are significantly more dangerous to pedestrians even when the EV drivers have the best technology.
Or that the 49% that aren't Teslas hit 6x as many people.
But seriously, there are actually some big problems with your assumptions here. First, the study includes hybrids, which aren't necessarily running on battery power, but which outnumbered Teslas substantially during the period in question.
Second, the study ran from 2013 through 2017. Tesla didn't even get Autopilot version 2 until 2016. Before that, it was using the MobilEye hardware suite, which had very limited capabilities compared with current versions.
Third, the Model 3 didn't come out until 2017. Tesla barely sold any cars prior to that year, so Teslas were not popular anywhere during the time period in questoin.
Fourth, the study was in the UK. By the end of 2017, Tesla had sold about 5,600 cars in all of the UK. Compared with hybrid vehicles, their sales would have been lost in the noise.
Fifth, the total number of plug-in hybrid and EV cars would have been small back then, so this data likely has a wide margin of error caused by hybrid and EV cars not necessarily being used in all the same environments where people drive gasoline-powered cars. For example, EV drivers are way more likely to be urban than average, which means that the average number of pedestrian encounters per EV is likely to be higher than the average number of pedestrian encounters per ICE car. I have no idea if they compensated for that.
In other words, the study doesn't really provide much interesting insight, IMO.
The pedestrian has one thing to focus on.
This says a lot about your perspective.
Pedestrians (and runners, cyclists, etc) are arguably focused on *more* things than drivers. They're moving much more slowly relative to cars, which means that every minute they're required to consider the potential dangerous action taken by a large number of drivers, from traffic in both directions.
Cyclists, yes. Pedestrians, no. Cyclists are in the worst of both worlds, moving at a speed fast enough for cars to not perceive them in the way that they would perceive pedestrians, but slow enough to have a higher risk of getting hit, and with limited protection against getting hit.
Pedestrians are not moving quickly, and have almost no trouble stopping and waiting for a few seconds to make sure it is safe to walk. At most non-metered intersections, that's all it takes is a few seconds. At metered intersections, pedestrians just have to wait for the walk light and not walk against the light.
Cars have a responsibility to pay attention to pedestrians when they are making turns, or when pedestrians are walking towards them in their lane on minor roads with no sidewalks.
But it is important to understand that every ounce of energy that drivers spend worrying about pedestrians is mental energy that they can't spend tracking vehicles that are moving much more quickly and thus can cause an accident much sooner. In relative terms, it is way, way, WAY safer for a pedestrian to pause on the sidewalk and see if it is safe to cross a road than for a driver moving at 35 MPH to pay attention to that pedestrian while potentially not noticing someone pulling out of a side street at a high speed.
No, it is absolutely NOT the case that drivers have fewer things to focus on than pedestrians. Pedestrians aren't moving. So per second, statistically speaking, pedestrians are tracking half as many moving objects as a driver who is moving at road speeds towards other cars moving at road speeds.
Every single car passing them has the potential to be piloted by a distracted driver that could easily kill them by both action or inaction. I have to stop to avoid being hit by inattentive drivers nearly every single time I go for a run.
Unless you're talking about jogging down a road with no sidewalks or waiting on the sidewalk before stepping out into a crosswalk because of someone who doesn't stop (which in most states they're not required to do, though they absolutely should if it is reasonable to do so), this sounds like a road design problem. You should petition your city to give pedestrians a separate crossing cycle at every intersection, and to have pedestrian crossing lights at every crosswalk.
Yeah, not sure what that test question was about on the driver's test. But even still, if you hit a pedestrian, no matter under which circumstances, I think the operator of the vehicle has quite a burden of proof on their side. Even if most of the "fault" is on the pedestrian, there is still going to be *some* fault for the driver as you should be prepared at all times to come to a complete stop.
Being prepared to stop is not the same thing as being able to stop in less than the vehicle's minimum stopping distance. No amount of preparedness can change the laws of physics.
I remember my UPS driver's training where they drilled into me that *every* accident is avoidable, 100%. As a matter of fact, they always told us that if we got into an accident, it's our fault until proven otherwise.
While that's nice from a corporation trying to avoid liability approach (blame the driver), it's nonsense.
People often point out that 97% of accidents are avoidable. While that may be true, that doesn't mean that 97% of accidents are avoidable by both parties. A large percentage of accidents cannot be avoided by the not-at-fault party without some sort of driver assistance (e.g. Tesla FSD doing emergency corrective steering to avoid someone who veers suddenly into your side).
I think it's more of what you think is more important - the machine or the human.
That's a false dichotomy, because it presumes that the only choices are killing the human or waiting for the human. There's a third option: the pedestrian waiting his or her turn.
Which makes more sense: having the pedestrian wait two seconds for the car to pass or having the car wait thirty seconds for the pedestrian to pass? I would argue that if all else is equal, the car should have priority, because the car will cause less time delay for the pedestrian than the pedestrian will cause for the car.
Of course, when car traffic is continuous, fairness breaks down unless you forcibly give pedestrians an opportunity to walk. But with your proposal, fairness completely breaks down with even a small number of pedestrians. Without time-slicing pedestrian traffic, it would only take about three or four pedestrians per minute to completely wipe out traffic flow through an intersection. That's why allowing pedestrians to have priority makes no sense.
The best way to solve that is to ensure that when pedestrian traffic is nontrivial, pedestrians are grouped together by a traffic light and all cross at once. That time slicing is why we have traffic lights that allow pedestrians to cross only at periodic time intervals. So basically, the status quo ensures some amount of fairness, and your proposal doesn't. That makes your proposal a huge step backwards. It should not be seriously considered.
So what *can* we *realistically* do to prevent collisions without destroying the utility of roads for cars? Give pedestrians their own crossing cycle. Allow pedestrians to cross in any direction during that cycle, and do not allow cars to enter the intersection during that cycle for any reason, including turning right on red. (Use a separate "No Turns" sign that lights up when it is illegal to enter the intersection.)
With that approach, pedestrians should never get hit at an intersection (more on non-intersections later) unless they flagrantly violate the law. And because the pedestrians can walk in any direction through the intersection, it saves them considerable time if they're crossing both ways, on average. And because the cars are never sharing the intersection with pedestrians, they never have to wait to turn while pedestrians block the intersection, so traffic also flows better.
This should be common sense. There are basically no downsides to this approach, and significant upsides. Everybody ends up getting where they are going faster with that approach, and it is safer, too. Yet for some reason, there are very, very few intersections that do this. And *that* is why so many pedestrians get hit by cars.
IMO the human takes precedence - darting out into traffic is something that happens and expecting the human to have presence of mind to note safe crossing points is the equivalent to subjugating people to machines.
At that point, driving will have to be so slow that you might as well walk. There's really no safe way to allow pedestrians to take precedence while still allowing vehicle traffic at much faster than a walking pace. Darting out into traffic shouldn't be something that "just happens", because that means that pedestrians aren't exercising a due standard of care, which puts them and drivers at risk, because they may be forced to dodge pedestrians by driving into oncoming traffic, light poles, other pedestrians or cyclists, etc.
That may seem like an extreme statement, but consider having to walk 1/4 mile in 100 deg heat just to get to a designated crossing point while the people in their cars have A/C.
If there are no cars coming, then that should not be illegal. BUT — and this is a big but — it is not reasonable for pedestrians to step out in front of a car, or walk without looking both ways to see if it is safe. It absolutely must be the pedestrian's responsibility to look out for him/herself. Everyone must share responsibility for road safety. Expecting cars to be solely responsible for pedestrian safety is naïve and foolish, and results in road systems that are utterly ineffective.
More importantly, though, if there are a large number of pedestrians that cross at a particular spot, then there should be a proper crosswalk with a warning light that doesn't involve walking a quarter of a mile. Ideally, that warning light should trigger automatically as pedestrians approach the road, giving drivers extra warning that they should slow down or stop. Responsibility doesn't end with the pedestrians or the drivers. Governments must also take responsibility for providing reasonable safety systems to make pedestrian crossing safer. And if they aren't doing that, it is your responsibility to vote for someone who will, or if necessary, run against them.
believe me when I first moved to Berkeley I almost couldn't drive trying to take into account all the bicycles and pedestrians but I got used to it and it's second nature now.
There are plenty of Bay Area streets where I refuse to drive, precisely for these reasons. It isn't worth dealing with roads that are clogged with pedestrians. In effect, those roads are nonfunctional. This includes large parts of San Francisco. That's not solving the problem. That's a copout. That's giving up and saying that the problem is too hard, so the only thing we can do is make driving miserable and hope people don't get mad enough to sue the city for the wasted fuel and the resulting catastrophic environmental impact of their poor road planning.
Kids dart into traffic all the time so it's the kids fault when that happens --- this is not a sane argument.
When I see kids near the side of the road, I tend to slow down just in case. However, it is still important to teach kids the importance of not running out in the street without looking. Any parent that doesn't do that is a fool. It isn't at all reasonable to normalize that sort of behavior. Even in the best of circumstances, stepping out in front of a car is like playing Russian Roulette. Maybe you'll get lucky, maybe you won't. Taking unnecessary risks for no good reason is always a bad idea.
And more importantly, changing the right-of-way rules still won't make it safe to do that. No rule changes short of a 5 MPH speed limit will make that safe. At real-world speeds, when it comes to a collision with a car, it doesn't matter to the pedestrian whether that pedestrian had the right of way. The pedestrian is still dead.
Do we really want to label the death of Adobe Flash as an "anticompetitive" thing? I say good riddance. Leave it in the past with old Geocities stuff like blink and marquee tags on websites.
Oh, don't get me wrong, I think that the death of Flash was a good thing, because it was a blight on the world in every respect, from performance to security, but it still demonstrates Apple's power to dramatically manipulate related markets, which makes it important from a legal perspective.
Ebooks: thanks for explaining.
Car: If I rip open my car my warranty is void so no I can't do that any differently than I can dual boot my car or iPhone.
No it isn't. Magnusson Moss et al ban such practices. If you break something while changing the radio, the warranty won't cover that damage, and they potentially won't warrant the radio if you're using it outside of your car, nor the new radio that they didn't sell you, but it does not affect the overall car warranty at all. By law.
My Spotify app uses their logo and brand name. Can't do that without cutting a deal with Spotify.
Sure. That's not the same thing as Spotify writing the app. And either way, that's largely moot for the other reasons mentioned below.
The same as me wanting to write my own car and iPhone app.
No, it isn't, because your car isn't a general-purpose computing device. It is potentially a safety-critical system, and either way, it is an embedded system.
And unlike my phone I have no way to install my own car app, nor does GM provide any sort of documentation or new app upload App Store site for me as a random developer to do so.
Because unlike your phone, your car is not a general-purpose computing device. The very fact that GM does not even create the illusion that everyone can develop apps for the platform makes it materially different from a legal perspective.
Very locked in to what GM has chosen for my car. Isn't that a worse monopoly control situation than iPhone?
More control, yes. Monopoly, no, because an embedded system isn't designed to run arbitrary software, doesn't have a marketplace for software, etc.
Additionally, even if it were a monopoly, there's no clear consumer harm from your car's software being locked down, and it isn't illegal to exercise monopoly control in ways that don't harm consumers. The critical difference between those two pieces of hardware is that the consumer impact from the car being locked down is approximately zero, whereas the consumer impact from the phone being locked down is potentially huge.
Most people carry a phone with them at all times. For anything that the phone can do, the user gets very, very little benefit from the car being able to duplicate that functionality. Therefore, as long as your car provides a network-based API for your phone to control various aspects of the car, your car doesn't *need* to do anything but be a minimal control system for your phone. Everything else that it does is a minor convenience feature.
By contrast, most people only carry around one phone, because carrying around more than one phone is a huge pain in the backside, not to mention a considerable expense — $$$ for the phone, plus $$ per month for the cell service. So if a phone can't do everything they need it to do because of arbitrary manufacturer restrictions, that's more than a minor inconvenience. It's a huge hassle.
Your car not doing something, therefore, doesn't meaningfully hinder your ability to do it, because at least 90% of people can just use their smartphones to do it. Your phone not doing something, however, does meaningfully hinder your ability to do it, because most people don't have a second smartphone or other mobile device that they carry around with them, and if they do (e.g. a smartwatch), they are usually made by the same manufacturer as their phones and are locked down similarly.
That's the consumer harm, and that's critical to establishing the illegality of monopolistic behavior.
Just because we use as single word to refer to human faculties collectively doesn't make them one thing (e.g. planning). Billions of years of evolution have furnished us with a swiss army knife of cognitive abilities from being able to proto-count (subitize) to being able to infer where other people around us are directing their attention.
A lot of our intelligent behavior is being able utilize these disparate capabilities *together*. For example, I notice the people around me are looking at some other people, who are looking back; I recognize that some of these people are in my in-group and other are outsiders; I perceive (subitize) that there are more of them than of us. There are some people who can't do this because they aren't as capable as other people at various links of the chain, and yet these people are often highly "intelligent" in other ways.
As we build AI tools, there is little point in them unless they *exceed* human abilities in some manner. So arguably we have already AI tools that, on certain tests with well-chosen constraints, are smarter than humans in a very narrow and specific way -- certainly in their ability to process large volumes of data. What we won't get at first is that kind of seamless integration of different kinds of mental capabilities. This integration is so natural and effortless for us we call all our highly disparate abilities by a single word.
My -guess- is the ebook market is highly fragmented so 10% was a lot but as I said, not familiar.
It's not. It's barely lost in the noise compared with Amazon. That particular case was about collusion to raise prices. The thing is, Apple's contracts with ~500,000 developers who have apps on their iOS App Store, if found to be behaving in a way that effectively raises prices, would make the size of the eBook price fixing cabal seem tiny by comparison. As soon as multiple companies and contracts get involved — which is absolutely the case when you're talking about a closed ecosystem like iOS — you don't actually have to have monopoly power to be convicted of Sherman Act violations at all. (And yet Apple does.)
My car actually does have apps. I can run things like Spotify, a weather app, and there's an App Store with more crap I haven't dug into but Spotify is definitely a third party.
Yes and no. Spotify is a third party, but the Spotify "app" in your car isn't necessarily a third-party app. Spotify has a public web API. Either way, the fact that it is just a small part of your car makes it an embedded platform, which genuinely is different from a general-purpose computing device.
Yet I can not dual boot my car to another carOS.
You didn't buy your car computer. You bought your car. You can rip out the entire head unit and stick in a different one from a different manufacturer if you want to do so. Nothing locks you to using that head unit. And that's actually an important distinction.
Additionally, the largest car manufacturer by U.S. sales volume has only 17% of the market, and most people connect their phones to their cars and use that for everything anyway, making the car's head unit basically a glorified Bluetooth speaker. Based on that, you'd have a *really* hard time convincing anyone that any of those companies have *any* meaningful market power over the app market no matter what they do in terms of the software on their head units. You'd also have a hard time convincing anyone that the power that they do have is meaningful, given how easily that apparent market power disappears just by connecting your phone.
Yes Apple has a 100% monopoly over iOS. But for that to be a problem iOS would have to be a market. It isn't.
Actually, if you read the court's opinion in U.S. v. Microsoft, arguably iOS *is* a market unto itself. But that's not really relevant, because at 60% of device sales, 68% of worldwide app revenue, and by my math, possibly as much as 91% of U.S. app revenue (iOS at 60%, 7x spending by iOS users, so 60*7 / ((60 * 7) + (40 * 1)) ==
It's a product. That's like complaining Ford has a 100% monopoly on Fords. Why can't I build and sell my own Fords with my own SmarterThanCarOS on it?
You can. In fact, there is or was a huge market for folks doing things like van conversions, where they buy Ford vans and massively rework the interior, complete with different head units, etc.
Besides that, Ford isn't a platform. It's a car. People don't buy a Ford for the user interface of the computer on the dashboard. They buy it because it's a car. The software is a minor part of the functionality, and as previously noted, is entirely replaceable without it ceasing to be a car. This is not true for an iPhone or iPad. And that is where the biggest distinction lies.
I know Apple's global share is much lower but I wasn't making that point.
Smaller than Ford, with its 14% U.S. market share and 5% world market share? Hardly. Apple's worldwide phone market share is equivalent to the worldwide of Toyota + Volkswagen + Honda + Hyundai + about half of Ford. (That's the top 5 put together, BTW.) But that is mostly moot, because you aren't buying a car dashboard computer. You're buying a car. And Ford doesn't prevent you from putting in a different radio, nor prevent you from using your phone to play content over the sound system.
Maybe that's because you still think market share is the sole determinant of whether a company has monopoly power. It isn't. Multiple court rulings have made it clear that there is no bright line at any particular number. Circumstances have to be taken into account, regardless of the specific numbers, though as far as I'm aware, no case has ever been successfully brought against a company with less than 50% market share unless it involved collusion with other companies that, when added up, totaled more than 50% market share (e.g. the eBook case that Apple lost when they had only 10% market share).
But the thing is, Apple is well above that line at ~60% market share. And in terms of app revenue, the iOS platform is even higher than that, both worldwide and in the U.S., putting it well above the threshold where it would typically be considered to have monopoly power over the market, and it has been that way for a very, very long time.
There are multiple factors that impact whether a company has monopoly power, including:
All of these things play a role in determining whether a company has monopoly power over a market. Apple has clear monopoly power over multiple markets related to mobile apps. And that's before you even start considering platform lock-in concerns, iMessage incompatibility, and other questionable behavior.
What I think the DOJ is going to attempt to show is a pattern of abuse — not just one action or a small number of actions, but a large number actions that cumulatively add up to consumer harm in the form of increased pricing for apps, in-app purchases, music, books, movies, streaming video services, and even physical devices, plus decreased security, decreased user choice, etc. And there's enough of a pattern here that Apple should be worried.
If you want to know where corruption lies, follow the money. Specifically, if the DOJ were to win this case, who gets the money? The taxpayers sure as hell don't get the money.
In a manner of speaking they do, because that money pays down the national debt a tiny bit. But more importantly, cases like this are intended to be punitive. If the company fails to stop doing whatever they're doing wrong, penalties can stack up rapidly until they do.
So no, this is not a shakedown. This is the U.S. government seeing the EU take similar actions against Apple, and deciding to follow suit in an effort to change Apple's behavior. All Apple has to do is stop acting like a monopolist, and all of this will go away.
Normally/traditionally 90% was the the minimum market share required to be a monopoly.
You know, depending on who you ask, you'll get 50%, 60%, or occasionally 70%. I've never seen anyone argue a number as high as 90% for the threshold, except perhaps in the context of an economics class, for the purposes of distinguishing a monopoly from an oligopoly.
I kind of like what Thomson Reuters has to say on the matter, which is that if you're over 70%, you're probably a monopolist, and if you're under 50%, you're probably not. For companies in between, it probably depends on the circumstances. And even those numbers are not hard rules, just a first approximation.
Good analysis.
Do you think a company can be hit as a monopolist when they only have 60% of the very competitive market? Normally 90% is a monopoly.
There is no hard limit for being declared to have monopoly power of a market. The general consensus is that it isn't typically used for companies with less than 50% market share, but Apple is above that threshold in the United States. I'm not sure where you get the 90% number, but that has no basis in the law. Even at 90%, a company can not be a monopoly if there are no real barriers to entry, and even at 50%, a company can be found to have monopoly power if the barrier to entry is high enough.
And either way, no particular number would be small enough to preclude charges under attempting to monopolize a market, though, which is also covered by that same section of the Sherman Act, should a company conspire in sufficiently interesting ways. For example, Apple is a convicted monopolist because of certain clauses in its iBooks Store terms of service, even though it had only a 10% market share in the eBook market.
So the short answer is yes, they absolutely can.
Market share is critical. Saying, "well they tried to be a monopolist" is insufficient to make them guilty of anything. For example, a company with 2% market share is trying soooo hard to be a monopolist. Uh, yeah, no one cares at 2%. DoJ wouldnt even blink in the 2% direction.
There's clearly a line at which it would be challenging to claim that someone is attempting to monopolize a market in the absence of conspiring with other companies. I don't think 52% of the U.S. cell phone market falls below that line, though. And the U.S. as a whole is a market for antitrust purposes, so there's no free lunch from "Yeah, but the rest of the world has more Android phones." When you're talking about a product that most people only own/use one of at a time, any company that is selling O(120 million) of that product every year, almost by definition, has a strong enough market position to be subject to antitrust scrutiny.
Additionally, one can reasonably argue that the App Store terms of service is per se conspiring with other companies (app vendors). And the DoJ actually did blink at Apple's role in the eBook price fixing scheme despite its ~10% share, so I wouldn't agree that the DoJ won't care at 2%. It all depends on the circumstances. Market share is not the primary factor, or even necessarily the most interesting factor.
Finally, iOS provides 68.13% (source) of worldwide app revenue, and the percentage of U.S. app revenue from iOS is probably even higher (because of the higher market share for the devices). That's *way* above the point at which it starts being interesting to the DoJ. Otherwise, they wouldn't be bringing this suit.
I don't agree with your split of the iPhone hardware vs software. Phones are not PCs and the court is unlikely to see them in the same light. If they had burned the os onto ROMs which would make the OS a function of hardware, there'd be nothing to look at. This is no different than me not being able to change the OS on my car. It's embedded.
First of all, most people *can* change the OS on their cars, though maybe not on the existing head units (and to be fair, who would want to reuse those bottom-of-the-barrel pieces of junk). There are plenty of third parties replacing the head units of cars, doing custom self-driving stacks, etc. And they can do that because with few exceptions, things aren't that locked down. The few companies that do lock things down (e.g. Tesla) are relatively minor players in the market.
Second, your car isn't designed to run third-party software. An iPhone is. Arguing that an iPhone is an embedded device is just plain silly. It doesn't meet any of the criteria for being an embedded device. It is an end-user device unto itself rather than being part of another device, it is designed to allow running arbitrary software created by companies other than the manufacturer, it has a faster CPU and more memory than many computers from just a few years back, etc.
The App Store is a different situation as it involves a market with third party developers, etc but I don't see how "I can't put a different OS on my iPhone" is going anywhere. There's nothing special about iPhone hardware. It's actually inferior in many respects to similar priced Android competitors which would allow a different OS install so no one is being hurt by the the OS lock in.
Well, that's where I disagree strongly. If an iPhone allowed you to run a different operating system, you could ostensibly dual boot into iOS and Android, and run whatever apps you want while under Android, and it wouldn't be under Apple's control. It is the combination of multiple layers of protection — the hardware not allowing you to run different operating systems and the operating system requiring all software to be sold through Apple that causes it to be problematic.
And there are multiple different ways in which it is problematic.
The above are all examples of Apple attempting to leverage its monopoly power over the iOS app/content market to monopolize other markets. The case for an antitrust action against Apple is, frankly, obvious. I cringed at the App Store rules back when I first heard them. My first thought was, "That's an antitrust case waiting to happen," because it was a very clear attempt to monopolize various markets. The only thing surprising is how long it took for any of those lawsuits to actually stick, given Apple's market share. But I guess they didn't feel comfortable bringing a case until Apple was pushing at 60%.
Seems to me Apple raised some significant legal questions in their initial letter. Those then become issues for the judge to address as a 'matter of law'.
IANAL, but seems to me Apple raised some very legitimate legal issues in their letter. I'd expect the judge would want to resolve those kinds of questions rather than being overturned on appeal.
I'm a bit concerned that their lawyers my have a very casual relationship with the truth. They didn't even make it through the first sentence without saying something that is flatly wrong as a matter of law:
"A Section 2 Sherman Act claim can move past the pleadings only if the complaint alleges (1) monopoly power in a relevant market;"
That is, of course, not even remotely true. The section of the legal code in question begins "Every person who shall monopolize, or attempt to monopolize, or combine or conspire with any other person or persons, to monopolize any part of the trade or commerce among the several States
It doesn't get any better from there:
"The Supreme Court has repeatedly held that the type of conduct at the core of this case—namely, Apple’s decisions about how and whether to grant third parties access to its platform—does not give rise to Section 2 liability as a matter of law."
Also no. More on this below.
"and the complaint fails to allege that Apple has the ability to charge supra-competitive prices or restrict output in the alleged smartphone markets"
... none of which is per se required for an antitrust claim.
They then proceed to cite two cases as precedent that are about phone companies and line sharing agreements, which while somewhat related to the antitrust concerns that might be raised by, for example, Spotify, are about limits on who those companies do business with, which is not material when the company is actively preventing other companies from doing business with the consumer directly. If Verizon or PacBell prevented, for example, Comcast from physically running lines to the houses of people in its territory unless they sold their houses and bought new houses somewhere else, the situation would perhaps be more directly applicable to a case about Apple preventing users from installing apps on their phones without selling those phones and buying phones made by another company.
The Novell v. Microsoft case is not even tangentially related; the primary antitrust claim was rejected because the statute of limitations had expired, leaving only a secondary claim that Microsoft hadn't made APIs available that Novell wanted, and the courts rightfully concluded that any harm to Novell's business was from dragging their heels at bringing their software to Windows 95, rather than because of those relatively minor APIs.
New York v. Meta Platforms is similar to the telecom cases, but more hilariously, was dismissed not because it was without merit, but because the state was prohibited under the doctrine of laches from even making the case after such a long period of time. And although this could theoretically apply to this case, that isn't the argument that Apple is making. It also isn't clear that litigation over a pattern of ongoing behavior would be estopped by laches — merely retroactive damages for behavior prior to a certain point — unless Apple can somehow prove that obeying the law in the future would cause irreparable harm. [insert laughter here]
They then try to distance themselves from United States v. Microsoft Corp. by pointing out that Microsoft had a 95% market share. While true, Apple's 52% market share in the U.S. isn't far enough away from that number for such a claim to hold up, IMO, and it seems highly likely that the court will agree.
They also say "Apple, by contrast, does not restrict third parties’ abilities to deal with competitors." Except that in the Microsoft case, the one small part of that case that fell into that category involved Microsoft restricting third-party hardware manufacturers from making other operating systems and apps available as an out-of-the-box experience. Apple effectively does *both* of those things, just a little bit differently.
In Apple's case, they make all iOS hardware themselves, so you can say that there isn't a conspiracy involved, but that isn't necessary to cause consumer harm. And Apple *does* *actively* prevent third-party operating systems from running on their hardware, which from a consumer perspective, is not meaningfully different. Ironically, by mentioning that, Apple runs the risk of the DOJ adding additional charges.
The same is true for Microsoft preventing OEMs from pre-installing Netscape on computers. Apple prevents anyone, including consumers, from installing Firefox (a real version, not the WebKit skin) on any hardware running iOS. The fact that Apple no longer licenses their operating systems to third parties, again, doesn't fundamentally change the character of the concern.
But the gut punch is this line from the U.S. v. Microsoft case:
"It is therefore proper to define a relevant market that excludes the Mac OS."
Using that standard, one could argue that Apple's lawyer's initial statement is, in fact, even more incorrect, because it is arguably proper to define a relevant market that excludes Android, Windows Mobile, etc.
The last part of that section basically comes down to a matter of opinion about whether they are showing a theory of harm. This is unlikely to fly, given that Apple is basically arguing that they should skip the part of the trial where such a theory would be demonstrated. Were this at the certiorari stage, it might be a reasonable argument, but as a pretrial motion, it falls flat.
Part II, again, is similar, in that they're arguing for skipping the actual arguments, and they seem to be conflating two different types of consumer harm — the harm from lock-in at a platform level with the harm from lock-in at the app level — both of which are presumably alleged.
Part III raises the same nonsense about monopoly power. If those apps existed in separate markets, you'd be able to buy them without going through Apple. The fact that more than half the smartphone users in the U.S. can't is per se evidence that their argument is at best laughable, and at worst, in bad faith.
Then, they again claim that Apple does not have monopoly power in the smartphone market, which brings us back to the point from U.S. v. Microsoft, and the little problem of Apple selling more smartphones in the U.S. than literally all of the other smartphone manufacturers combined. What's the threshold in their minds at which a company has monopoly power?
And finally, in the very last paragraph, they finally mention attempted monopolization as an afterthought, almost as if one of the authors suddenly realized at the last minute, "Oh crap, there's this other huge part of that statute that we completely ignored for the whole letter." Intent to monopolize can be shown through a pattern of monopolistic behavior, which Apple has, IMO, exhibited on an ongoing basis for at least the last decade, and possibly longer. So again, this seems to come down to an attempt to keep the government from actually presenting its arguments. And again, if this were certiorari, they might have a chance, but not for a pretrial motion.
In short, I see nothing in that letter that is, at least in my mind, a "very legitimate legal issue". It's the usual sort of pretrial motion that Apple files in response to every lawsuit against them, and they rarely have much luck with them. I expect this one to be no different in that regard.
You're not Dave. Who are you?
