Apple is a State
Why Apple is the internet’s first state—and why it challenges so many of our notions of a free, open, and interoperable web
It is, why not, 1974, and I’m trying to get you to invest in my firm. Stubbing out my cigarette, I make the pitch—there’s this one formula that has revolutionized finance, I explain, “The Black-Scholes Model.” I whisper its name as if it were some incomprehensible divinity because for me, your bushy-mustachioed fund manager, it is: I have no idea why it works, but I know it’s been perfectly predicting options prices so well that it doesn’t matter if its assumptions are wrong. Everyone believes in it, so it’s started becoming true. And dig it, there is a lot of money to be made from this formula whenever options are mispriced.
What I don’t tell you is that this is only half the story, and what I’m really curious about is your data. I want data on where you live, I want data on how much you’re willing to spend with me, I want data on how your spending increases with each successful trade, I want data on how often you follow my recommendations, and I want data on how well this Black-Scholes thing really works. I do not see myself as unscrupulous. Nixon is about to resign for wiretapping, Ralph Nader beat General Motors in court a few years ago for spying on him, and J. Edgar Hoover is two years dead. Those cases represented intrusions on privacy. I’m just asking you to give me your data willingly so I can sell it to the Computer Men, who promise they’ll eventually give me some business insights by correlating these things. A year ago, in 1973, The US Department of Health Education and Welfare (HEW) warned against the implications of computers on surveillance—“an individual must increasingly give information about himself to large and relatively faceless institutions, for handling and use by strangers unknown, unseen, and, all too frequently, unresponsive”—but I’m not worried. I tell myself that I am subject to the American State only as much as it stipulates that I am, and there are no data laws to stop me.
Still, it is clear I am quite subjected to the American State: subject to its 48% corporate taxes, and subject to its surveillance, as Nixon and Hoover revealed, even as my own ability to engage in surveillance will be increasingly curtailed over the coming decade. By 1977’s Whalen vs. Roe, wherein the government outlines protections on informational privacy, it will be clear I’m subject to the American State’s provisions on data collection as well (luckily, so will the American State). Even if I somehow realize just how prescient HEW’s warning really is, the US’s incredible power over my life likely makes me think that America will someday circumscribe the powers of computers—but that the chances of computers ever circumscribing the power of the US are very slim. That is to say: it is only 1974, and with the sounds of Come and Get Your Love wafting from nearby car radios, I am fairly ignorant of my role in creating a new world order.
Because, now, it is 2021. And what I’m selling you today is an app—really, a doozy of an app that I’ve proudly titled PumpTruck. Trying to contain my glee, I scour your eyes for inevitable validation as I explain the glory that is PumpTruck: it lets you join pools to pump stocks that only activate the trades when enough users have joined. You seem less excited than I think you clearly should be, so I push further: it lets you pump any stonk you like bro, and it’s legal! In truth, PumpTruck might serve any of a handful of similarly noxious functions, but the legal is the major point here. For what I’m most proud of is how I’ve evaded the state in ways that would have been difficult 50 years before. I’ve figured out that pump-and-dumps are legal if crowdsourced, and I’ve turned that into a viable product with off-shore accounting to avoid all taxes. In 2021, the American government can’t touch this.
There is just one issue. I still have to pay a 30% fee to Apple’s app store, and I won’t be able to collect much data on users to improve my recommendations, not because the American government prevents me, but because Apple requires your consent to allow me to track data even as it harvests this data for its own targeted ads.
And here we both pause, recognizing the dim echoes from half-a-century prior. The limits on my own surveillance by a centralized institution that implements its own, the infrastructure to enable my business, the taxes it takes as a consequence—all of this sounds strangely familiar from my mustachioed days selling you options.
Apple is now our state.
This likely sounds ludicrous. Even if we accept that for a tech business in 2021, Apple serves the same role as the American government did in 1974, we might object on plenty of other grounds to the notion of Apple as a state in the first place. It doesn’t force anyone to live in its regime, doesn’t use violence to coerce its user-inhabitants or competitor-enemies against behavior it deems bad, and it doesn’t manage its own monetary policy. But to what degree are any of those qualities necessary to qualify as a state, particularly a state on the internet that can’t have recourse to physical violence and must operate globally with all currencies? What if, instead, we defined states more simply, using a proposal from James C. Scott’s Seeing Like a State? “Society became an object that the state might manage and transform with a view toward perfecting it.” A state, now, is an entity that implements infrastructure for its population in order to manage their lives for the supposed better—regulating companies for supposed protection of its constituents, even as it exempts itself from these same regulations.
And this is exactly what Apple has done. Apple has arguably become the first state on the internet, usurping the place of the American government and EU in developing infrastructure that not only enables businesses to thrive, but regulates and sucks the value from them in return. It would be tempting to say the same of, say, AWS or Stripe. But AWS and Stripe do not represent constituents. Only Apple, as an internet state, represents a citizenry loyal to its kingdom and forces businesses to follow its laws in order to court that populace. This is also what separates Apple from not-quite-states like Facebook and Google: businesses advertise in Facebook’s and Google’s kingdom, but they don’t build their products and apps to operate inside its codes. (The possible exception here is Android, but Android often operates as a knock-off App Store for businesses to redeploy the apps they built for Apple.)
I say kingdom, but Apple is, more precisely, a fiefdom: an estate controlled by a centralized overlord (the Apple Corporation) that offers small parcels of its operation (apps) to vassals (developers) in exchange for fealty, or more exactly, fees from farming the land (reaping money from Apple’s users). At the heart of recent complaints against Apple is the question of whether we really want the world’s biggest company to adopt a model of 14th century feudalism that turns its competitors into its vassals and its users into harvestable land. For example, in “Apple is Holding Back the Creator Economy,” Li Jin, Nathan Baschez, and Yash Bagal list five grievances, all indicative of Apple’s fief-like operation: payments must transact through Apple’s purchase system, must incur Apple’s 30% cut, can’t surpass $999, have to be disputed and refunded through Apple as an intermediary, and may not even happen if Apple decides arbitrarily that it doesn’t like an app. “Imagine the cry if Google Chrome were to only allow payments using Google Pay,” they write. In an equally wonderful piece that inspired this one, Matthew Ball digs into the implications of that gate-keeper ecosystem: consumers lose because companies have to charge more, companies that submit to Apple’s payments lose because they lose their profits to Apple (on every $100 of revenue, Roblox loses $25 while Apple makes $30), companies that skirt Apple’s billing system lose because the Apple alternative becomes easier-to use (Apple Music vs Spotify), and Apple wins because it can favor its own apps while extracting massive value and data from its competitors.
For those of us devoted to the future of Web 3.0, it’s tempting to say that Apple’s consolidation of power is what should weaken it as well. Its subpar apps and services—from Apple TV to Safari to Maps—survive only with the state-sanctioned support of its superior products (phone, computer), and its refusal to make its apps interoperable with other publishers and operating systems puts Apple Music, for example, at a significant disadvantage to Spotify, as Baschetz argued recently. Ball makes the case that Apple’s fiefdom has positioned itself directly against the nascent metaverse, which will be built on “interoperable standards, device/endpoint agnosticity,” the free exchange of items and currencies across apps, and “marketplaces for labor, creation, and consumption that that disintermediate Apple.” Ball echoes one of the most forceful claims against Apple, Ben Thompson’s argument in December that Apple directly violates the spirit and currents of 21st century technology. “Apple is systematically destroying the ability of platform-driven small businesses to compete with the Internet giants,” writes Thompson, with “anti-competitive” policies “predicated on control of the App Store, and a demonstrated willingness to use its power to get its way.”
That means, for Thompson, that Apple is violating the principles of a free and open internet: “Computers emit data as a matter of course, and the Internet makes the transfer of that data free. To strive for a world without the generation or capture of data is to fight against the very nature of technology.”
And yet, Thompson and Ball’s utopianism about the future of a free-and-open internet—which I very much share—masks a more unsettling historical record. Apple has become the first state on the internet not in spite of its closed architecture but because of it.
Indeed, go back decades, and you’ll see complaints lodged against Apple’s “closed architecture”—that is, its suite of systems and services that only operate on its own devices. Even as Apple gradually builds its empire, these complaints sound a note of protest remarkably like those of the past year, repeatedly that Apple’s authoritarian control of its system should disincentivize creators and thus users in turn. Consider this note in the Cider Press from 1985:
“Because of its closed architecture, the Mac draws much less support from the entrepreneurs of the world. Apple's general attitude to outside developers is one of selective cooperation. Many potential ventures with Apple are headed off at the gulch. While Apple is large enough to survive with this attitude, it is giving up valuable markets to others and causing loyal users to desert it. There are signs that this strategy is going to backfire on Apple unless they open up their architecture to the user world.”
Or take, more recently, this charge from Peter Cripps in 2011:
“It’s one of life’s great anomalies as to why Apple is so successful in building products with closed architectures when most everyone would agree that open architectures and systems are ultimately the way to go as, in the end, they lead to greater innovation, wider-usage and, presumably, more profit for those involved.”
And so, to understand how Apple alone has been able to create a state—to draw and patrol borders on a closed fiefdom—it’s necessary to understand the ways that Apple has resisted this narrative of an open internet and interoperable computing for nearly 40 years. We can jump back to 1985. As recounted in Apple Confidential 2.0, Microsoft had recently copied much of Apple’s operating system for Windows, and it knew Apple could sue. Bill Gates, likely aware of Apple’s slow sales at the time, decided to press his advantage by threatening not to develop Word and Excel for Apple unless Apple gave Microsoft rights to use Mac technology for its own operating system. It was a deal Apple would regret for the next decade, and we might see it as the origin story for the Apple ever-since, which continues to operate its systems and services exclusively on its devices and will never license its software for outside use. It is a mode very much at odds with Microsoft and Google’s aims to make their own operating systems operable on nearly any device.
But why did Apple’s closed architecture, opposed to the open ethos of computing and the internet for a full 40 years, give it unfair advantage to become the internet’s first state? Thompson himself has given one very good reason, arguing just this week that Apple’s sandbox architecture and strong-arm control of the App Store created a malware-free environment for users to download apps that incentivized app development in turn. Note how much this sounds like a traditional argument for states to regulate business and keep streets safe in order to stimulate a healthy economy: Apple’s complete control of its ecosystem, from operating system to app store to payment processing, allowed it to protect its users and give trust to developers in a virtuous feedback loop for all. At a time when conducting downloads and payments on the internet was a decidedly risky proposition, Apple cleared the driftwood, kept out the varmin, and set up a town by appointing itself sheriff. This is what it means to become a state.
Still, we might offer any number of other reasons as well. We might say that the internet still needs to be accessed through a physical good produced through complicated supply chains that Apple has masterfully mastered back here in the physical world (i.e. the benefits of beautiful infrastructure). Or we might simply argue that Apple’s architecture isn’t really closed, but selectively closed: others can play in its sandbox under its rules, but it will never play in theirs. This, perhaps, is the key point of the Apple origin story in its refusal to license software. Unlike Android and Windows, Apple apps and services will only work on Apple products, and that is, in many ways, the source of its market advantage: Apple showcases, manages, and monetizes the services of its competitors, even as its own services are never showcased, managed, or monetized by its competition. (I.e. the benefits of a “Dutch hegemon”).
Nevertheless, I’d like to end by offering two other reasons for the success of a closed ecosystem on an open internet.
The first, simply, is tribalism. At a time when “brands are dead,” consider the cult-like devotion of Apple fans, a tribalism that is impossible to imagine ever translating to Facebook, Google, or Amazon, all of whom provide a plethora of fairly unintegrated services that we might use one-or-two at a time. By contrast, Apple’s closed architecture doesn’t just reward us for investing heavily in a family of devices by giving us exclusive access to integrated products and services—it tells us that we have joined a rarified tribe of blue messages against a sea of drabber green. And while the high quality of Apple’s products is a precondition for this tribalism, we’d be unlikely to feel it without Apple’s closed ecosystem reminding us constantly of the special access and privileges we get.
Indeed, under the guise of representing a counter-culture demographic of creators opposed to hierarchical power systems, Apple has always appealed to users to join its innately superior class of wealthy and genius leaders opposed to a working-class of corporate drones: you’ll find that ethos in Apple’s 1984 campaign, in its Think Different campaign, and its Mac vs. PC ads. What else does Apple’s high price tags signify except that we have paid to join this superior class—that we have, effectively, paid not only for a better product, but for status? And so, we might consider if the reason we really want a free and open internet is because it allows us to project our own superior status online more widely.
But the second is that in many ways, it’s easier to build an authoritarian state online than it is in physical reality. If we return to the central insight of Seeing Like a State, that the function of a contemporary state is to make society more legible as data for its own purposes (taxes, surveillance), then it’s clear that Apple is better at being a state than its nation-counterparts back on earth. After all, if the job of the state is to turn the messiness of nature and human relations into codified data in order to shape and extract value from relations, then an internet state has a much easier task, for it is dealing with a world created from data in the first place. It is simply easier for an internet state to survey items and behaviors, just as it is impossible for users to hide—there can be no evading taxes from the Apple State as there might from the US Government.
Of course, in the future, this distinction may become garbled as physical states increasingly become online states as well, leveraging internet surveillance and data collection for their own traditional ends. In fact, we might end by speculating that Apple’s state is in many ways quite similar to China’s—that is, a closed architecture state whose significant investments in online infrastructure for citizens and businesses allow it to control both. China, of course, is not a fiefdom, just as Apple is not run through “fragmented authoritarianism.” But with a bit of poetic license, we could extend the comparison in order to define just what a closed architecture state means: for example, that it doesn’t attempt to operate imperialistically within competing regimes (as the US and Microsoft have historically done in order to manage other states/firms from within) but does impose significant barriers-to-entry to outsiders while making dangerously jingoistic suggestions about the superiority of its population. Closed architecture shouldn’t be mistaken with isolationism, which has become the domain of America; rather, it happily allows outsiders to enter, as long as they play by the rules. When we talked about how Apple manages its competitors, but is never managed by them—to a great degree, the same is true of China as well.
Ultimately, the incredible success of Apple and China—perhaps the most powerful company and country in the world, both leveraging their control over the internet—present formidable challenges to any framework we might have of a globalist future where ecosystems freely interoperate without regard for borders or centralized authority. I remain hopeful that the critics are right, and that what worked for Apple in the past will fail in a decentralized, user-generated future that allows people everywhere to build their own worlds and implement their own immutable laws through smart contracts and DAOs. Likewise, I remain hopeful that Brian Lui’s predictions of China facing upside decay will apply to Apple as well as it disincentivizes creators to develop for its platform.
But standing here as a citizen of AppleLand, realizing the ways that Apple has leveraged the powers of the internet to build an empire in which I play a tiny part, I also worry—that at the dawn of Web 3.0, with critics the upcoming opportunities of interoperability in an internet metaverse, Apple is not only an internet state, but got that way by arguing and proving against all nay-sayers that the internet was never meant to be free.
***
The argument above came out of a fervent dialog with Sidhartha Jha, to whom I am extremely grateful.