Hacker News Reader

Comment by _qulr | Hacker News Reader

Scale and market power are what makes a monopolist, by definition. It doesn't matter if the terms have changed; what has changed since 2008 is the scale and power of Apple.

Incidentally, the 30% came from the iTunes Music Store, on which the App Store was based 100%.

The most popular game consoles have 2-3 thousands games total. They are truly "curated". The Apple App Store and Google Play store each have 2-3 million apps each. They're orders of magnitude larger than game consoles. This is the crucial difference in platforms.

View on HN

Apparently the 30% cut has it's origins in a deal between Nintendo and Hudsonsoft. 10% licensing fee, and 20% manufacturing costs for physical cartridges.

https://www.msn.com/en-us/news/technology/the-30percent-fee-...

If Steam, Apple, Google, any of them made a physical cartridge of a developer's games for them, that would seem a lot more reasonable a fee.

What about what the virtual platform Apple provides? The cartridge is the feature rich phone and virtual services Apple provides to get to the device.

For which a customer spends hundreds of dollars, comparable to purchasing a game console that took cartridges. Apple doesn't provide that, buyers provide that. Apple is trying use their leverage to dominate every angle of the action. Surprise, surprise, surpriiiiise.

And millions of developers pay $99/year, which likely more than covers Apple's costs of providing developer tooling, documentation, etc.

I wouldn’t be so sure. How many man hours go into developing ARKit, Foundation, Swift etc.? How many developer accounts would be required to fund all of that plus the R&D for the ideas that didn’t work out?

I think the $99 is more of a filter, like a captcha, for developers that mean business.

The user paid for those when they bought the phone.

Probably a more accurate analog to the cartridge would be the storage and distribution services. That takes some infrastructure in networking and storage.

Developers still have to store application specific data on their own!

Apple actually gives developers 1PB of free storage per app with CloudKit, and it even has some nice privacy features for users. Of course it has limitations, so most developers prefer rolling their own backend.

How's that? I choose apps that store my data locally rather than their servers if I even do download apps.

I was discussing the storage and distribution infrastructure.

Consider a game that holds a leaderboard on steam, or any app that stores user progress on its servers.

Not exactly pertinent to your comment but pertinent to the conversation as a whole: If Apple/Google/Steam stored all app data for developers and also did the marketing for developers, that is closer to what 30% is worth imho. For hosting the platform, yes your grandfather was the first to stab a flag into the digital real-estate, and for that I owe you 30%? Feudalism that manages to shift cost to both consumer and developer at the same time -- impressive, but not very noble.

That's not a good analogue for a cartridge, though. The cartridge or the device itself stored data locally. Modern gaming consoles store their data on HDD's. Anything beyond that is entirely developer's prerogative.

And according to another user, they do offer storage through CloudKit... which it appears they do, included in the dev account/App Store distribution model—1 PB: https://developer.apple.com/icloud/cloudkit/

It's not really the same because the reason Valve built out all of this infrastructure was so they could take the middleman's cut. Basically, GameStop buys games cheaper wholesale and resells them at a higher retail price. Valve gets all of that by being the direct retailer to the end consumer.

Not even having to manufacture a physical good is a separate cherry on top of that.

This is a super literal take. In the 80s/90s the expense-incurring activity was physical cartridge production. Today it’s bandwidth & delivery infrastructure, card processing, any myriad of things.

Yes, but it's hard to imagine that 30 years of technological progress has not changed the underlying cost structures of these businesses. After all, it's changed the cost structure of almost every other business that has some connection to computing and data delivery by orders of magnitude. Yet, strangely the 30% cut has not changed a bit.

NES carts used to cost like fifty bucks. Aren't most apps in the few-dollars range?

Nintendo console games still cost $50, and seem to outsell $1-$5 games despite the fact that the vast majority of users already own a phone, and consoles cost extra (and there are more than two platforms to choose from.) So I'm going to go out on a limb and suggest that the market does not consider these to be equivalent products.

The still costs $50 is interesting to me, because inflation adjusted this is a pretty steep discount. The 30% stayed the same, but inflation adjusted on something like this the revenue netted is down to what 15% would have previously got you.

All I’m really getting at is that the layers of complication are thick. I feel like one could unwind this with a lot of effort, but I doubt anyone is being utterly fleeced (or is utterly innocent, for that matter).

> This is a super literal take. In the 80s/90s the expense-incurring activity was physical cartridge production. Today it’s bandwidth & delivery infrastructure, card processing, any myriad of things.

Are the current-day costs of those "myriad things" in any way comparable in magnitude to the 80s cost of cartridge production? I highly doubt it. Bandwidth is cheap enough that single webpages are several megabytes each, if you load everything. Card processing fees are 3-4%. My bet the breakdown of Apple's 30% is >25% "licensing" and <5% cost of provided services.

The cartridges are actually kinda a funny story to themselves if memory serves. Nintendo required that they manufacturer the cartridges themselves (and they had minimum orders), which was separate from the licensing side of things. Then you had disk based game systems come along early 90s which were a couple of bucks to manufacturer (on avg). I remember reading about wild deltas depending on your company size & how much you shipped on what you could expect your take to be on a title, which also makes it hard to compare.

The is the wrong history though. It only matters if you accept the App Store == console analogy, which I don't. Historically speaking, App Store was not based on game consoles, it was based on the iTunes Music Store. Everything about the App Store design was exactly the same as the iTunes Music Store, right down to being in iTunes itself. App developers even uploaded their apps to iTunes Connect (which has only very recently been renamed to App Store Connect).

That’s a little like saying since the facts of how the people at Apple arrived at the original 30% disagree with your philosophical views on this matter, it seizes to be relevant. A personalized reality, with personalized facts. Alleged fact, much as it may not make sense to some in hindsight, is they used the Nintendo-Hudson Soft price model and it doesn’t change how they decided that because of anyone’s preferences. It’s totally different than to accept, it’s been this way - for too long - it’s time to change for a more equitable price model. But then that’s the crux about bringing this suit to court in a country whose judicial system is based on precedents.

> Rogers has adjudicated various cases against Apple. In 2012, Rogers dismissed a class action lawsuit with prejudice, upholding Apple's defense that the "Illinois Brick doctrine" from the Supreme Court case Illinois Brick Co. v. Illinois, 431 U.S. 720 (1977) applied, as only the developers of apps could be damaged by Apple's policies, and consumers did not have statutory standing to bring suit on the developers' behalf. The Court specifically noted that the 30% fee Apple collects is "a cost passed-on to consumers by independent software developers".[8] The United States Court of Appeals for the Ninth Circuit reversed her decision, and the Court of Appeals was upheld by the Supreme Court in Apple Inc. v. Pepper.[9]

> In December 2014, Rogers presided over a jury trial against Apple, in which plaintiffs claimed DRM on Apple iTunes violated antitrust laws. On December 16, 2014, the jury reached a verdict in favor of Apple.[10]

Source: https://en.m.wikipedia.org/wiki/Yvonne_Gonzalez_Rogers

> The Court specifically noted that the 30% fee Apple collects is "a cost passed-on to consumers by independent software developers".

This is a somewhat shocking conclusion to come to since the independent developers neither set the fee, nor process the payments. I can see why it was overturned.

At the same price of a cartridge today they could give you the app + a basic smartphone

> Scale and market power are what makes a monopolist, by definition.

Very true. Worth noting here also is that (US) antitrust law isn't designed to break up 100% of all monopolies.

US antitrust law isn't designed to break up platform monopolies, period. Which is unfortunate, because the kind of all-owning combinations it's designed around aren't really possible in tech, post-Internet.

Microsoft of the 90s was much-lamented, but in reality they extracted far less value than they could have. They could have easily asked for 30% of all revenue of all Windows software, and developers would have had to pay it.

I'm hesitant to advocate breaking up of platform monopolies or duopolies, but a lesser-version of antitrust law would substantially benefit consumers in both app stores and ISPs. In both cases, the monopoly stems from monopolized access to the end customer, which is what would be beneficial to attack.

IMHO, a sliding scale of mandates for increasingly-open third party access based on market share (broken down into each user market!) would be appropriate.

If you want 90% market share, go for it. But you'd better believe you're going to have to provide (1) customer access to third parties & (2) any work required to enable them, at cost.

Just a reminder that actually by definition consumer harm is the bedrock standard for antitrust litigation in the US. Scale and market power may be components of an argument for consumer harm, but they are not sufficient in and of themselves.

> consumer harm is the bedrock standard for antitrust litigation in the US.

Which is really dangerous, IMHO, because a lot of the most vile bits of monopoly power can happen without directly causing immediate consumer harm.

You shouldn't need to prove consumer harm to win antitrust litigation. That should be just one of a few different harms, any one of which should be able to win antitrust litigation.

Can you give examples of other kinds of harm you are thinking of?

To ask you the same question, restated, that the judge asked Epic, which they had no answer for: if Apple wasn't a monopoly before, but they're a monopoly now, when did they become a monopoly?

While it is an interesting question, what does it matter? When did standard oil become a monopoly? Is there some specific date where they changed from competitive to anti-competitive? More likely they grew over time and the ability for competition to exist was diminished over the entire period.

Apple doesn’t even have 60% market share in the US. Android has 86% worldwide. They are no where near a monopoly like Standard Oil or Bell.

> Apple doesn’t even have 60% market share in the US.

They have over 50% market share though. They're by far the largest smartphone manufacturer in the US. And they also get the vast majority of app developer revenue, much more than the Google Play Store.

> Android has 86% worldwide.

Worldwide market share is not necessarily relevant to US law.

> They are no where near a monopoly like Standard Oil or Bell.

This isn't actually required by antitrust law.

Worldwide share could be made to matter during the trial (assuming it reaches that point). It demonstrates that Apple's position in the US is a consequence of consumer choice, as they do have real competition. It's just that their competitors have (for various reasons) failed to take (or in this case failed to hold) a larger share of the US market.

https://www.statista.com/statistics/266572/market-share-held...

Shows the marketshare of mobile OSes by year from 2012-2019. It's worth noting that Apple's > 50% market share is a recent (2019) phenomenon. It's not like they've held that position for the entire decade, and it's not at all clear that they got that position by any illegal or unethical methods [0]. So consumers have made a clear and (somewhat) informed decision by electing to by iPhones.

[0] I'm sure someone will talk about green bubble/blue bubble, but seriously people: stop using SMS/MMS it's crap. Get your friends onto Signal or, hell, even WhatsApp so you can have real cross-platform, secure (less some metadata) communication.

Apple and Android together have a duopoly or an oligarchy over mobile phones.

Apple also has a lock over it's own platform for which switching costs are very high.

If people switched phones every few weeks, this would be less of an issue.

What is the high switching cost?

Hardly anyone buys apps. They either pay for services that work cross platform and the majority of sales are coming from in app consumables for games.

Media? iTunes music you buy has been DRM free for a decade and Apple Music is available on Android.

Movies? Blame that on the studios that don’t participate in Movies Anywhere. Apple is a member along with Amazon, Vudu, and Google. If you buy a movie from a participating studio, it syncs between all of the platforms.

Books? No one buys books from Apple.

"What is the high switching cost?"

People buy phones once every year or two, they're about $1K and data is controlled by the providers = High switching cost.

Your favourite Marmalade ... were it to double in price next week at the grocery store, could be substituted without any friction.

The ASP of an iPhone is $675. An Android phone is less than $300. That doesn’t take into account the ability to sell your phone and/or give it to someone.

What do you mean “data is controlled by your provider”? You just switch the SIM card.

ASP is meaningless because who had iPhone won't buy dirt cheap Android phone.

The ASP of an iPhone is around $700

But now are you saying that we need government intervention because of “lock in” when people are using their own free will to buy more expensive products when their are cheaper more “open” products available? Are people too dumb to make their own choices.

Data isn't locked to one platform.

You can transfer photos, mail, browser history etc between phones and app data is of course managed by the app.

Standard oil never had more than 20% marketshare.

This question assumes that Standard Oil didn’t engage in anti-competitive practices after they gained a monopoly position which they absolutely did. On the run / don’t have time to send links but this is common knowledge and easy to look up. It’s worth noting that a monopoly alone (a definition for % market share) does not meet the criteria for antitrust. Antitrust carries an extra burden of the company abusing its monopoly position eg what Standard Oil did and arguably, what Apple is not doing.

You seem to be asking your second question rhetorically, but its actually an interesting question. The reason we call anti-monopolistic activity "anti-trust" is because Standard Oil organized itself as a series of companies related only through a common trust. That trust controlled 40 different companies, and was the original target of the government's ire. When the trust was ordered dissolved, it eventually was replaced by a holding company.

The history is interesting and informative, and explains a lot about why the US government started to care about issues of corporate governance in the first place. When people talk today about the dangers of "monopoly," they often misunderstand the point entirely.

Size alone is not the issue. It never has been.

The reason it matters is because Epic's entire theory of the case involves Apple having a "monopoly" over the specific sub-market of "iOS app distribution". They've intentionally defined this market very narrowly because they know Apple does not have an actual monopoly in the smartphone market.

As the judge pointed out yesterday, Epic has "created a failsafe definition". Because they have defined a single product market, Apple by definition has a monopoly over their own product.

But assuming Epic's theory is true, it would have been true from the very start of the App Store, back when Apple had barely any market share. Of course, it would have been absurd to complain in 2008 that Apple was unlawfully maintaining its monopoly over "iOS app distribution", which is why Epic's lawyer largely dodged the question during the hearing yesterday.

Epic asserts they’re a Monopoly but that’s a contentious assertion so the easy answer is they aren’t.

Not a direct answer to your question, but somewhat related. When the iPhone was first released (or a year later when the App Store was released) we did have more competition than we do today.

Windows Mobile, Symbian OS, BlackBerry OS (and to some extent Maemo) all had the opportunity to compete with iOS and Android but were either too slow to transition to market demands or did not get a lot of traction to start with.

While not officially a monopoly in my book, I think it's safe to state that we're now down to iOS and Android for 99% of the market with KaiOS Ubuntu Touch (Tizen?) picking up some of the more niche or cheap devices.

If lowering your price doesn't increase demand then you are in a way a monopoly. I don't see any scenario at all where Apple lower their cut from 30% except via government action, so it makes sense to legally force them to either get competition or to lower their cut below 30% to something more sensible.

Or maybe your addressable market is saturated....

Either you are a monopoly for that market or lowering prices will let you gain more market share. More market share is an incentive so there is a reason to reduce prices. You could probably find some reasonable exceptions, but nevertheless I think if a company has no incentive to reduce prices then it probably should be regulated.

Why is “market share” an incentive to reduce prices? Companies don’t chase market share. They want profit. Market share doesn’t pay bills.

There is roughly 50% of the US population who wouldn’t buy an Android phone if it cost $1.

Having incentives to do something doesn't mean that you will do it. Lowering prices reduces your income per sale, that is a disincentive. If it increases sales then that is an incentive. If there are no incentives then there is a problem since then there will never be a reason to reduce price.

Healthy competition is when the different parts have incentives to reduce prices to steal customers from each other. Then unless they collude they will reduce prices until there is barely any profits left. However if reducing prices doesn't help you get the other persons customer then there isn't healthy competition.

So tell me, what non government action would you say could cause Apple to reduce their 30% fee? I don't see any. Apple reducing their 30% fee wouldn't cause more people to choose Apple over Android and vice versa under current circumstances. Both of those fees are set with no care about competition.

> There is roughly 50% of the US population who wouldn’t buy an Android phone if it cost $1.

Pretty sure many would switch if the highest end android phones were free.

Apple has been in business for over 40 years.

- in the early 80s, Apple //e’s were more expensive than the competition.

- in the mid 80s - 2000, the Macs were and continue to be more expensive than the equivalent PCs

- in the early 2000s, iPods were more expensive than the equivalent media players.

- the iPhones are more expensive than equivalent Android phones.

Apple hasn’t “found a reason” to reduce prices in 4 decades. Did it have a “monopoly” on computers, music players, tablets, watches, monitors, etc all that time?

> Pretty sure many would switch if the highest end android phones were free.

“high end” Android phones are cheaper than iPhones. Yet and still Apple dominates the high end. Maybe every one doesn’t buy solely on price.

Well actually, there is no Android phone that performs as well as the $399 iPhone SE, but that’s another debate....

I don't think you quite understand.

The person you are responding to basically quoted the definition of a monopoly, as according to the opinions of the US government.

If changing prices does not change demand for your product, then by definition, according to the US government, you have significant market power (IE, you have a monopoly. The definition of a monopoly is merely having significant market power, according to anti-trust law)

That is the definition that judges and the government use for market power.

As usual, it just takes a little observation to see how non sensical this interpretation is.

Do you really think that every company that can raise its prices is automatically defined as a monopoly? What next? Nike has a “monopoly” on sneakers because it can raise its prices on Air Jordan’s and people still buy them?

Should Apple also be regulated as a monopoly because it raised its prices on Mac Pros and people still bought them?

Could it possibly be that a bunch of posters on HN don’t know as much about the law as ... a real judge who didn’t agree with similar arguments.

If it was an alive frog before, but it is a boiled frog now, when did it become a boiled frog?

I'm no expert on frogs or boiling, but I would say the moment the frog is 100c all the way through. There seems to be a defined thing that makes it a boiled frog, but that doesn't seem to be the situation in the Epic v Apple case.

It very much matters if the terms changed, considering that the long shot legal precedent Epic is relying on is Kodak v. Image Technical Services. Terms changing in that case was a key element of why a single brand could be considered a valid antitrust market.

Could you explain this in more detail? I read about this case, and my understanding is that the Independent Service Organizations ("ISOs") didn't actually have a direct contractual relationship with Kodak. They were truly independent, and thus there were no terms to change.

Kodak monopolized the aftermarket for parts and service after their customers were "locked in." The primary market for their equipment was competitive and they didn't have monopoly power, but the Supreme Court still decided against them. This established that a single brand could be a "relevant market" for antitrust purposes under certain circumstances. This is the case that Epic is relying on against Apple. Unfortunately for them plaintiffs almost never win under Kodak antitrust theories.

> Kodak monopolized the aftermarket for parts and service after their customers were "locked in."

It's not clear to me that the timing was essential. Apparently Kodak had a policy since 1975 to only sell parts to direct purchasers of equipment, but the initial case wasn't filed until 1988. So the official terms, which were between Kodak and customers, not Kodak and ISOs, hadn't changed recently.

Clearly Kodak was taking some non-official anticompetitive actions, but that's not the same as changing the official terms.

Subsequent decisions by circuit courts have used the lack of an explicit policy change to find against the plaintiff. This case is at the outer boundaries of antitrust law and lower courts have effectively taken the perspective of the dissent to limit its applicability.

> lower courts have effectively taken the perspective of the dissent to limit its applicability

The dissent... of the Supreme Court opinion??

The "terms" in this context are not referring to contractual terms between Kodak and the ISOs, they are the conditions under which Kodak customers could obtain service and repair parts for the copiers they originally purchased.

In Kodak, customers originally purchased Kodak copiers without any aftermarket restrictions, and then several years later Kodak changed their policy and stopped selling repair parts to the ISOs, which left customers locked into buying repair services from Kodak instead.

The key factor in Kodak was that there was change to the conditions under which customers could obtain repair parts that customers could not have reasonably anticipated or planned for when they originally purchased Kodak copiers. The fact that a change in policy occurred is very important to that case, and indeed some subsequent courts have found that absent that change in policy, Kodak does not apply.

Epic's lawyer spent the first part of yesterday's hearing attempting to argue that Kodak applied to this case, but judge seemed very skeptical of this line of argument. As Apple's lawyer correctly pointed out, single-brand markets are a unicorn in anti-trust law and this case does not fit the narrow exceptions created by Kodak (and a related case, Newcal).

How do you weigh the value of Apple's APIs?

As in, to continue with the game console analogy, you must have at least one license to a powerful engine (like Unreal Engine with a 5% fee) to develop efficiently.

Do you actually save 5% (assuming PS is 30%) by developing for the Apple ecosystem? (Assuming Metal is comparable to Unreal)

> Do you actually save 5% (assuming PS is 30%) by developing for the Apple ecosystem?

Probably not - assuming you're being smart and targeting multiple platforms, you're still going to be using Unreal/Unity. And even if you're only targeting the Mac, Unreal/Unity bring a lot more than rendering to the table, so you're probably still using either of them, all with a commercial license.

And unless you're doing distribution yourself, anywhere you put your game will take a percentage, pretty much all around 30%. Some places like itch.io let you define how much they take, but no one really uses itch.io for anything bigger than an indie game.

> And even if you're only targeting the Mac, Unreal/Unity bring a lot more than rendering to the table, so you're probably still using either of them, all with a commercial license.

Depends, plenty of indies are doing alright with Apple's own gaming frameworks.

Very true. Comes down to if you already know (or want to learn) Swift.

>Scale and market power are what makes a monopolist, by definition. It doesn't matter if the terms have changed; what has changed since 2008 is the scale and power of Apple.

Anti-trust laws have specific conditions to meet such as predatory pricing, product tying, exclusive dealing, etc. It's not just a matter of market share and market power, they have to have obtained or maintained that market position unfairly. This sets a pretty high bar in a market with 2.5 major app platforms (Apple, Google and Amazon).

Reply on news.ycombinator.com