Last week, in the midst of the slow, painful collapse of the generative AI hype cycle, something incredible happened.
On Monday, a Federal Judge delivered a crushing ruling in the multi-year-long antitrust case filed against Google by the Department of Justice. In 300-pages of dense legal text, Judge Amit Mehta confirmed in excruciating detail the length and breadth of what all of us kind-of-sort-of-knew — that Google has a monopoly in search and online “general text advertising.”
The ruling precisely explains how Google managed to limit competition and choice in the search and ad markets. Documents obtained through discovery revealed the eye-watering amounts Google paid to Samsung ($8 billion over four years) and Apple ($20 billion in 2022 alone) to remain the default search engine on their devices, as well as Mozilla (around $500 million a year), which (despite being an organization that I genuinely admire, and that does a lot of cool stuff technologically) is largely dependent on Google’s cash to remain afloat.
Judge Mehta found that Google violated the Sherman Act — the century-old antitrust legislation that, among other things, led to the break-up of Standard Oil in 1910, as well as the break-up of AT&T in 1982.
This is a significant moment in Silicon Valley history, and an existential threat to Google, which owes its existence to multiple monopolies and a mob-like approach to dominating the search engine market. Over half of Google’s yearly revenue comes from Google Search — over $175 billion — and it’s hard to imagine that it could sustain said revenue (or its dominance) without paying billions of dollars to starve the competition.
Moreover, this lack of competition is why the Google Search experience has become so utterly awful. While monopolies can feel a little monolithic (I’ll get to the complexities here in a little bit), understanding the damage they do is fairly simple: if you’re not worried about the competition, why would you worry about losing customers?
The incentives of the monopolist are inherently degenerative. The documents I covered back in April in The Man Who Killed Google Search are a great example, showing how Prabhakar Raghavan, the former head of Ads at Google, demanded an “increase in queries” — meaning more searches rather than better results — as a means of increasing the amount of time one spent on Google, which in turn allowed Google to show you more ads.
Monopolies are a big part of why everything feels like it stopped working. Competition isn’t just healthy, but vitally important. However you feel about capitalism, a better economy is one where there is actual, real competition between businesses based on their ability to meet customers’ needs. A company that fears the competition is one that will compete to win a customer’s heart, and conversely a company that has no real fear of losing customers — because they’ve dominated a market with a monopoly or set up a healthy cartel where newcomers are chased out — can make their product worse without fear that customers might do something annoying like “choose another option.”
You see it a lot outside of tech, too. Ticketmaster and its associated properties have dominated live events to the point that they control both on-sale tickets and the resale market, leading to things like “platinum seats” where Ticketmaster reserves tickets so that it can sell them for a higher price to desperate fans, which is why The Department of Justice sued it in May for its monopoly.
Four companies control 80% of US air travel after decades of acquisitions, allowing them to effectively charge whatever they want — and because there are so few of them, they happily keep their prices in line with each other.
Shitty internet speed, or terrible customer service? That’s because most Americans have no real choice in internet providers thanks to the monopolies of companies like Comcast, Charter, CenturyLink and Cox, who can charge what they want, perform at a fairly mediocre standard, and effectively abandon less-profitable rural areas, discarding some to ultra-slow DSL internet at an inflated price from scumbags like Frontier Communications, which is partly-owned by private equity firm Cerberus Capital Management, which owns part of Albertson’s — a grocery store that’s currently in the midst of a battle with the FTC over larger chain Kroger’s trying to buy the company, with the FTC (correctly) arguing that it would eliminate competition and allow one firm to control grocery prices.
Sidenote: Broadband infrastructure is incredibly expensive to build, especially when talking about a country that’s on the scale of the United States, where the population density (37 people per square kilometer) is far lower than, say, the UK (279 people), France (122 people), or Germany (243 people).
Realistically, unless you have deep pockets — I’m talking about hundreds of millions, and potentially billions of dollars in funding — it would be hard to replace the incumbent internet providers. And only a handful of institutions have the means to actually do that.One — ironically, given the subject matter of this newsletter —is Google, which operates its own FttP (fiber-to-the-premises) network in nineteen cities, delivering speeds as fast as 8 Gbit/s.
Cities are the other option, and municipal fiber — where the town or the region acts as its own broadband provider — has become a popular, effective, and affordable alternative to companies like Cox and Comcast. And obviously, these companies have handled the rise of municipal broadband well, using it as a motivator to improve their prices and their services.
Just kidding, they fucking hate it. Every so often, a new tax-exempt lobby group crawls out of the woodwork to run misleading scare ads that paint municipal broadband as wasteful or, in the case of Utah’s UTOPIA Fiber network, a “dystopia.” These lobby groups are often structured in a way that means they don’t need to disclose their donor identities — though anyone with a working cerebellum knows they’re funded by the big ISPs.
Occasionally, these ISPs will cut out the middleman and give generous campaign contributions — which is a comforting euphemism for “bribe” — to politicians that will advance anti-municipal broadband legislation. As a result, sixteen states have legislation designed to prevent or slow the introduction of affordable alternatives.
These states tend to skew red (like Texas and Alabama) or purple (like Michigan and Pennsylvania). And yet, ironically, some of the best examples of municipal broadband are found in staunchly Republican areas like Chattanooga, Tennessee, which had full fiber-to-the-premises in early 2010, and has since emerged as an unlikely hub of tech activity beyond the usual suspects of New York, San Francisco, and Austin.
Remember when “inflation” raised prices everywhere? It’s because the increasingly-dwindling amount of competition in many consumer goods companies allowed them to all raise their prices, gouging consumers in a way that should have had someone sent to jail rather than make $19 million for bleeding Americans dry.
It’s also much, much easier for a tech company to establish one, because they often do so nestled in their own platforms, making them a little harder to pull apart. One can easily say “if you own all the grocery stores in an area that means you can control prices of groceries,” but it’s a little harder to point at the problem with the tech industry, because said monopolies are new, and different, yet mostly come down to owning, on some level, both the customer and those selling to the customer. There are no alternatives.
This is a fairly basic part of all commerce — one cannot have a lawyer represent both sides, and one cannot have the same agent represent both the buyer and seller of a house, for example.It’s also something that antitrust legislation has, from the very beginning, been concerned with. In the late 1880s, the 38 biggest train companies in the US joined forces and established the Terminal Railroad Association — an entity that acquired bridges, rail lines, and other essential infrastructure around the hub city of St Louis — for the sole benefit of its members. New member companies could only join the Terminal Railroad Association with the assent of all existing members, which was never granted. This tactic wasn’t just “vertical integration,” but a way to kill future competition, and the Supreme Court ruled the Terminal Railroad Association was an illegal monopoly in 1931. While train companies could own and operate infrastructure (itself a conflict of interest), they couldn’t do so in a way that was partial, or could restrict competition by imposing unfair terms on access or membership.
And yet, it’s these conflicts of interest that the tech industry has made billions off, and what I’d argue is incapable of operating without.
Meta’s monopoly over the advertising on both Instagram and Facebook (as well as its black box algorithms) allows it to make the products worse to show you more ads, and increase advertising prices whenever it wants — something it’s done multiple times without anybody really covering it. As a result, Meta is also the only source of truth of how successful your advertising is — to the point that it tricked the entire media industry into “pivoting to video” using phoney numbers.
Meta — as with any digital advertising monopolist — can basically say whatever it wants to you, inflating stats where it needs to as a means of getting more money, which is why there’s a $7 billion class action suit against the company for doing exactly that.
Again, it’s about incentives. If Meta has no competition, Meta doesn’t have to act honestly, or give detailed metrics about ad performance, because there’s no other game in town.
Apple has — at least, outside Europe, where the European Commission has ruled the company must allow third-party app stores — the monopoly over the iOS and other associated app stores, meaning that it can set the prices (such as its maligned 70/30 revenue split), what can be installed, and how said apps are monetized — which is why the Department of Justice recently sued it for antitrust violations. There is only one way to buy apps on your iPhone, and that’s why the App Store is so utterly swamped with microtransaction-pumped filth and random flashlight apps that feature invasive advertising — because there’s no quality control when you don’t have to compete.
If there were alternative app stores, Apple would be more incentivized to make the app store a better experience. At the very least, it would be able to incentivize the creation of better — less exploitative, less invasive — apps reducing the commision it skims from the revenue of developers.
Indeed, Apple being the only advertising firm on the app store is…also a problem. Apple has aggressively moved against tracking of iOS users — a good thing! — yet holds a complete monopoly on any advertisements on the iOS app store. You’ll just have to trust them, as it’s the only way to advertise apps to American users. Apple is also being sued by the DOJ for alleged Sherman Act violations over allegations its control of services like iMessage — which is unavailable on Android or Windows, and as internal documents show, is seen as a useful tool for ensuring customer loyalty — has allowed it to restrict competition in the messaging, smartphone, and wearable markets.
Competition is good. You want competition. Every one of these companies has sold you a lie — that allowing them to control the experience makes it “better” because it’ll be “curated” by the company that made it. Yet the companies that make these things are huge, constantly lay people off (though this isn’t the case with Apple), and when given the chance will take as many liberties as they can to squeeze as many dollars out of every user.
And when they don’t have to compete, they get lazy. They get domesticated. They don’t have to innovate or impress you, they just have to keep you there — and keeping you there can be as simple as “it’s really difficult to move” or “where else are you going to go?” I keep using my iPhone because I genuinely like iOS, but I also know that if iMessage was an app rather than a thing only found in Apple Products, I’d absolutely consider using another platform.
Social networks are, in and of themselves, monopolies, and the simplest way they maintain them is by making it harder to find your friends or posts on other platforms. Even then, once somebody has gotten used to posting one place, it’s hard to move them to another, because part of the product is the monopoly — the platform itself is where your friends are and the conversation is. The Fediverse is a nice idea, but the implementation requires the platforms themselves to agree and for individual users to agree, which is… a pain in the ass that’s going to preclude most people.
Facebook is a good example of this. When it launched, it was an objectively better platform than MySpace. And yet it took until May 2009 — nearly five years after it launched — to surpass MySpace in monthly US user numbers. The network effect is real, and it’s not something that can be washed away overnight, but rather chipped away over the course of several years.
One of the best pieces of advice I got in the last few months was to try and replace the word “growth” with “monopoly” when I’m writing, something I won’t necessarily do but will be significantly more aware of. The growth-at-all-costs Rot Economy is fueled by the crude oil of the monopolist, with Google bankrolling its other enterprises with its search monopoly, Meta with its advertising monopoly, Amazon its its ecommerce (and now logistics and labor) monopoly, and Microsoft, to a much lesser extent, with its dominance of the desktop productivity software market — though its had its own antitrust war that I’ll get to later.
Monopolies are inherently anti-consumer and anti-innovation, and the big push toward generative AI is a blatant attempt to create another monopoly — the dominance of Large Language Models owned by Microsoft, Amazon, Google and Meta. While this might seem like a competitive marketplace, because these models all require incredibly large amounts of cloud compute and cash to both train and maintain, most companies can’t really compete at scale.
No private startup can actually afford to build and scale a Large Language Model without the help of Google, Amazon or Microsoft — and thus they would, assuming they actually became useful, effectively control every single product involving this technology. While I genuinely believe generative AI is (in most cases) a product with minimal utility, minimal value, and will likely prove to be a transient fad, I’m genuinely terrified about the alternative, as it would turn a handful of already-successful tech companies into parasitic rent seekers skimming a cut from every app and service that uses a LLM.
And these companies are, in some cases, well-positioned to become tech’s next landlord. OpenAI, to quote Satya Nadella in an interview from November 2023, gave Microsoft “all the IP rights and all the capability [to its technology]” as a condition of its multi-billion dollar investment deal.
To be clear, I don’t think this monopoly (or oligopoly) actually worked because generative AI loses so much more money than it makes and has yet to solve the complex problems to justify its costs, but I hypothesize that the goal was (is?) for these companies to basically reduce the future of computing to you paying them for their models rather than you running your own models on their cloud compute services.
All that messy, annoying overhead of you “doing your own stuff” — like thinking “how much does this actually cost to run” and “could I do it cheaper?” — goes away because you can’t possibly hope to train a model as big or complex as theirs. You don’t have the money, or the power — after all, they own the cloud compute you need in most cases — to train your own models at the scale they can. Microsoft, Google, Amazon and Meta also monopolize the infrastructure itself, both in the data centers they can build (or already own) and in their access to the specialized GPUs necessary to run these models. Why would you even start building a competitor?
Sidenote: The DOJ is currently investigating whether Nvidia — which dominates the GPU market for AI applications — has violated any antitrust laws by imposing conditions that discourage customers from buying chips from AMD or any other company, in part by providing favorable prices on networking gear to those who buy its GPUs.
Additionally, the DOJ is investigating the company’s $700m acquisition of Run:AI, which provides GPU management software for AI customers, which closed in April of this year.
This is the ideal situation for a monopolist — you pay them money for a service and it runs without you knowing how it does so, which in turn means that you have no way of building your own version. This master plan only falls apart when the “thing” that needs to be trained using hardware that they monopolize doesn’t actually provide the business returns that they need to justify its existence.
When you have a marketplace entirely dominated by one company selling advertising on services it runs, you really have no idea what you’re buying. Take Google’s AI-powered “performance max” platform — a four-year-old black box advertising solution that covers all Google properties, letting you run one campaign across every Google property — for example.
Just tell Google what you want, how much you want to spend, and roughly who you’d like to see what you’re advertising, and Google will magically tell you at the end how many people saw your ads, and how much you owe Google for the pleasure — and no, you do not know where the ads ran, a problem that means that hundreds of large brands find themselves spending money on spammy and scammy websites that likely never get them any real customers.
This, by the way, is what keeps the lights on for these companies. Google and Meta own both the products and the advertising services available on those products, and thus set the terms of every part of the transaction, because there is no other way to advertise on Google Search, Facebook or Instagram. There are entities competing for placement, but that competition is done on terms set by the platforms themselves, which in turn control how and when ads are shown and how much you’ll be charged for them.
Crucially, they also control all the data involved. They are both the company selling you the ad space and the company auditing the ad space. If you do not like “Google” as an advertising vendor for Google Search, or “Meta” as a partner for Facebook, you’re shit out of luck — both if you want to try an alternative and if you need them to improve the product in question. This is the exact reason that Google is so nervous about its upcoming antitrust trial over its accused monopoly over the digital advertising industry — because the government wants it to divest its ad management division, which would in turn create an entirely different marketplace, one where it didn’t set the terms.
I also do not believe most of these companies can operate without their monopolies. Facebook and Instagram are increasingly-decaying products that make tens of billions of dollars a year because there is no competitor that can undercut (or compete with) Meta’s advertising dollars, which make up 98% of Meta’s revenue. Meta made $13.46 billion in profit last quarter on $39.07 billion in revenue, which is a direct result of it being able to manipulate the platform to show more advertising and increase the pricing of the advertising being shown. Meta’s users are trapped, because Meta owns two of the largest social networks in the world (three, if you count Threads, though it doesn’t show ads there yet), and Meta’s advertisers need to be able to access the swaths of people that the platform has. Everybody loses.
The same goes for Google, which made a profit of $23.62 billion last quarter on revenues of $84.74 billion. Do you really think that $61 billion in expenditures is sustainable in the event that Google no longer owns a monopoly over its advertising space? While those costs aren’t all associated with Google Search, half of the revenue is — and how much of said revenue is tied up in the vagueness of the metrics that it provides to advertisers?
Using the figures above, we see Meta had expenditures of $25.54 billion over the last quarter. Again, can it sustain this level of spending without a monopoly on advertising on Facebook and Instagram?
While we might like the main revenue-driving products and services from Apple (iPhones, Macbooks, etc.) and Microsoft (Xbox, Surface, Office 365), it’s hard to say the same of Meta and Google, and I believe both are deeply brittle as a result. A lack of competition has turned Meta’s core products into skinner-boxes that live only as a result of a combination of a black box advertising product and a product that billions of people rely on (and use for free) — much like Google Search and associated products like Gmail and Google Docs (which, I add, I do really like). These companies are not competing so much as they’re holding the digital lives of their customers hostage and using that as a means to set the terms of how they’re shown advertising that they control both the placement and price of. They are not worried about somebody using “the other” Facebook, or working with an alternative advertiser. Google doesn’t have to worry about losing you to another search engine, nor do it worry about you leaving Gmail, or Google Docs, or any of its products, because leaving is hard, and advertisers simply do not have another choice.
As I’ve written before, big tech has run out of hyper-growth markets to sell into, leaving them with further iterations of whatever products they’re selling you today, which is a huge problem when big tech is only really built to rest on its laurels. Apple, Microsoft and Amazon have at least been smart enough to not totally destroy their own products, but Meta and Google have done the opposite, using every opportunity to squeeze as much revenue out of every corner, making escape difficult for the customer and impossible for those selling to them.
And without something new — and no, generative AI is not the answer — they really don’t have a way to keep growing, and in the case of Meta and Google, may not have a way to sustain their companies past the next decade. These companies are not built to compete because they don’t have to, and if they’re ever faced with a force that requires them to do good stuff that people like or win a customer’s love, I’m not sure they even know what that looks like.
As I wrote a few months ago, Meta’s traffic is dropping. How does Meta win users back without savaging its revenue? If Google has to totally reconfigure itself as a result of one or two antitrust losses, how does Google unrig the dice of decades of monopolist rot? Are these businesses even sustainable if they can’t own both sides of the marketplace? These are, of course, all hypothetical questions, as we really have no idea what happens next. But I think it’s time that we — as consumers, and anyone reading this who’s a reporter — begin to evaluate these companies a little bit differently. Every single piece about these companies should evaluate how much competition they have and report whether or not they are using anti-competitive means to sustain their dominance.
You wanna see change? This is how we do it. When you write about Facebook’s revenue, it should carry an asterisk on it like Barry Bonds’ record breaking baseball (except Bonds was a great player without steroids, but that’s another newsletter), and any questions of Mark Zuckerberg or Sundar Pichai should actively and aggressively police their monopolistic tendencies — and the same goes for effectively any major tech company, private or public.
At the very least, I’d love somebody to ask Mark Zuckerberg how he feels about the idea of an alternative advertising network on Meta. He’ll hate the idea, just as Sundar Pichai would hate the idea of any kind of competition with Google Search or Google Ads. They must squirm — we must make them squirm — and we need to be talking about these monopolies every single day as loudly as possible so that they get more scared and either start doing something to change things or get their asses kicked by the government.
And I admit I’m a little excited about the latter due to, well, history. I’m not saying it’s impossible that things change — and indeed very much hope that they do — but Judge Mehta didn’t impose any punishment on Google. Said punishment will come at the end of an upcoming trial (separate to the one I just mentioned), which is solely intended to determine what remedies and penalties are appropriate.
And Google, no doubt, will fight this ruling to the very end, using every possible avenue of appeal until they’re either successful or run out of options. It will use every trick in the book — looking for procedural errors, or errors in Mehta’s interpretation of the law — to overturn the ruling, or to water down any punishment.
And I’m a little worried it might work.
Sure, there’s now a bipartisan dislike of “big tech” — meaning that it can’t count on the pro-business sentiments of a politically-appointed appellate judge, or the Supreme Court, which (following the appointments of Kavanaugh, Gorsuch, and Coney Barrett during the Trump Administration) decidedly skews conservative. Plus, the ruling — which, despite being longer than a novel, is worth reading — was forensic in its case against Google. The facts on the ground, and those obtained through discovery, are hard to argue.
On the flip side, Google has near-unlimited funds to throw at this legal battle — which, as I’ll point out later, is absolutely existential for Google — and for lobbying to try and change its fate.
And, depressingly, the US hasn’t proven particularly effective at curbing the cancerous expanse of big tech through antitrust legislation in the past.
And now for a little history lesson — and a story about how remarkably different the computing market could look today if said legislation had succeeded.
In the 1980s, MS-DOS rapidly emerged as the dominant operating system for IBM-compatible computers, and by the end of the decade, it held 80 percent of the market. While its growth was, in no small part, due to Microsoft’s chummy relationship with IBM, the company also benefited from changing market conditions, some savvy tech and business moves, and a few missteps by their competitors.
By the end of the decade, the computer market started to consolidate. As computer usage grew — at home, but especially in the enterprise — so too did the complexity of the operating systems, and with each version release, MS-DOS did more and more. Competing DOS products, many of whom were built with (comparatively speaking) skeleton teams, couldn’t keep up.
For context, Microsoft had 1,153 employees in 1986, whereas Digital Research — which made the competing DR-DOS — had around 240. Whereas Microsoft was already a multi-billion dollar company in 1991, that year Digital Research sold itself to Novell for a mere $80m.
During that decade, Microsoft started offering productivity software — like Microsoft Word and Excel — and started directly competing with the likes of Lotus, gradually building up its market share in those arenas. The release of Windows 1.0 in 1985 also started the company on a trajectory that would make computing — at least, in theory, as the first couple of versions of Windows were not particularly good — more personal and accessible to a wider audience.
But it wouldn’t be fair to ascribe Microsoft’s success solely to its vast resources and its technical acumen. It also resorted to stronging-arm hardware manufacturers into only selling computers with Windows/MS-DOS, or making Windows/MS-DOS artificially more preferable than buying an alternative.
One tactic — which started in 1983 and continued until the mid-1990s — was to force manufacturers to sell an MS-DOS license with each machine, even if the customer wanted something like BeOS or OS/2. This would be done by strong-arming vendors into buying licenses in bulk numbers — and not, as you’d perhaps expect, by the demand for MS-DOS or Windows.
How did it do this? By setting the individual license price for MS-DOS so high that buying in bulk was a necessity, and outright refuse to sell to any vendor that purchased alternatives — or refuse to sell them Windows. Here’s a great paper about it from the Summer 1995 issue of the Antitrust Bulletin Journal.
Microsoft’s insistence on bulk licensing agreements meant that asking for a competing operating system wouldn’t actually reduce the cost of the computer, but rather increase it, because the vendor has already bought a specific number of licenses and they’re incentivized to bundle the cost into each machine they sell. Buying something like DR-DOS would, in fact, mean paying twice for an operating system.
These licenses were typically only valid for two years, and Microsoft didn’t provide refunds for unused licenses, meaning vendors had an incentive to aggressively offload them onto customers, like by providing free or cheap MS-DOS licenses to anyone who bought a new computer, or by framing MS-DOS as a perk of purchase.
Another tactic — designed to bolster sales of its productivity software, while crippling its rivals — was to use secret APIs to control Windows that were only accessible to its own software, and not to its competitors. Microsoft’s rivals claimed that these public APIs were, generally speaking, not as good, and meant that their software performed worse on Windows, which by the early 1990s, had reached near-monopoly status in the PC market. The YouTube channel RetroBytes goes into this practice in some detail, if you’re curious.
Anyway, my point is that Microsoft was a bit of a bastard back then, and by the mid-90s, it was already facing some scrutiny from the FTC, which opened investigations and delivered a couple of tepid slaps on the wrist — which Microsoft, of course, fought.
In 1995, Microsoft released Internet Explorer (IE) — its first entrant into the web browser market, which competed with early stalwarts like Mosaic, as well as Netscape, which was the dominant player, and Opera, which was rapidly gaining momentum.
Microsoft knew the Internet was going to be a huge deal, and so it did everything it could into making Internet Explorer the dominant browser. This included bundling IE with Windows, and structuring IE so that it couldn’t be removed without breaking parts of the operating system. Rivals also claimed that Internet Explorer benefited from hidden APIs, which, in turn, meant that competing browsers wouldn’t work quite as well.
In 1998, the US Department of Justice filed suit against Microsoft, accusing it of breaching the Sherman Act — the same law that Judge Amit Mehta ruled Google violated last Monday — by establishing a near monopoly on the x86 computer market.
And Microsoft lost.
The entire episode (and the release of Bill Gates’ deposition tapes, where he responded with “I don’t recall” to most questions, and asked for the definition of words like “we” and “ask”) was a humiliating episode for Microsoft. In 2000, Judge Thomas Penfield Jackson ordered the breakup of Microsoft into two components — one responsible for the operating system, and one that would create applications.
Microsoft, as you’d expect, appealed, and the following year an appellate court overturned the ruling on a technicality, as Judge Jackson had improperly discussed the case with the media before its conclusion. While the findings stood, Jackson’s ruling was overturned, forcing Microsoft to reach a settlement with the Department of Justice.
And so, we got another slap on the wrist. Microsoft would have to share its private APIs with competitors, and agree to five years of monitoring, where a Department of Justice representative could, unannounced, show up to its offices and demand access to source code and records. The Department of Justice also created a technical committee — which consisted of three experts, none of whom could be recent Microsoft employees or contractors — to ensure its compliance.
Microsoft could, however, keep bundling software like Internet Explorer with the operating system — which meant that its near-monopoly on the browser market would remain unchallenged for another eight years, until Google Chrome arrived.
So, what happens next?
There’s four likely outcomes here, each with their own advantages or disadvantages, and their own level of probability of coming to pass. And hat tip to Adam Kovacevich for highlighting these.
Google may be ordered to cease engaging in default search deals, leaving the likes of Apple, Samsung, and Mozilla free to sign agreements with other search providers — although, these agreements would almost certainly be less favorable than their current agreements.
Judge Mehta might — although the ‘how’ is difficult here, as Mozilla and Apple aren’t defendants in this case — force browser developers to offer users a ballot screen, where people could choose what search engine they’d like to use as a default. This would probably boost the market share of smaller search providers by a small degree, although I imagine most would just opt — either through habit or preference — pick Google. At least, until a viable alternative gains momentum.
Google may be forced to share click and query data with rivals, although again, this poses its own thorny issues — particularly when it comes to privacy, though this is an obvious competitive advantage that Google holds.
Finally, Judge Mehta may order the break up of Google, compelling the company to divest Chrome, Android, or even its advertising division. This remedy is arguably the most potent of the bunch, and the one Google is most likely to oppose.
Chrome and Android are based on open-source products — the Chromium Project and the Android Open Source Project, respectively — and short of prohibiting Google from contributing code or providing funding, it would have a leading role in their developments, and thus, influence. Moreover, Google’s presence in Android isn’t in the operating itself, but everything else — called Google Mobile Services, or GMS.
GMS is the package of software that turns an Android phone into a Google Phone. It includes the apps that come pre-installed when you buy a new Samsung or Xiaomi phone, like Gmail, Google Chrome, Google Assistant, and so on, as well as the APIs that allow third-party applications to integrate with them. Any ruling that doesn’t force Google to divest these components is, for the most part, toothless. And any ruling that says, effectively, that Google can no longer build apps — or, rather, build apps for the mobile OS it’s most closely associated with — is hard to imagine.
Moreover, Google could just ditch Android. It’s working on its own OS, called Fuchsia, and it’s already using it in production devices, though it’s unclear how invested they truly are in the project. Combine this with native support for existing Android apps — which it likely already has, and if not, could be implemented as others, namely Microsoft had (albeit briefly), and Huawei pretended to — and it would effectively skirt any restrictions imposed on Google’s involvement in Android proper.
The biggest, scariest thing would be any separation of church and state between Google Search and its advertising platform. Doing so would be potentially ruinous to the company (which is why Google will do basically anything to stop it), but it’s the remedy that would actually change things. Google’s lack of competition in the search engine industry is a symptom of its ability to set the pricing and terms of advertising on it, a problem that can be solved by removing its ability to do so.
The question, of course, is whether Google Search can actually work without a monopoly. As I wrote in my last newsletter, search is ungodly expensive and has to constantly scale to keep up with demand, even in its rotten form — and that likely costs a great deal of money that it wouldn’t be making if it wasn’t the ones controlling both the amount that ads cost and the data around their success.
In fact, one of my chaos hypotheses is that this might actually be Google’s argument — that Google Search has become “too big to fail,” and thus requires such a monopoly to keep it alive. One might say this is a “skill issue” and “not my god damn problem,” and they’d be right. Indeed this might be a problem of its own creation. If other search engines were capable of competing with Google, it might not hold such a heavy burden.
As I’ve said, whatever Judge Mehta rules, Google will fight it, and it’ll play dirty. And the thing I fear is that Google will be able to use its massive hordes of capital and scare-tactics around data security to hound the government into a settlement that doesn’t meaningfully address the monopoly.
Even if Judge Mehta delivers a crushing blow to Google, which the company fails to successfully challenge, it’ll try to figure out ways to live with the ruling while maintaining its monopoly. Apple is a master at this.
When the European Commission forced it to allow alternative rendering agents (at the moment, browsers like Brave and Google Chrome must use the underlying technology behind Safari, meaning they’re unable to differentiate themselves in things like speed and the adoption of new web standards), Apple created a set of rules that — while technically allowing for alternative rendering agents — were so onerous, nobody will actually bother.
Similarly, when the European Commission forced Apple to open up app distribution to other companies, it responded by a set of rules that forced alternative providers to hand over 17 percent of their revenue. This, in essence, was designed to make it hard for alternatives to compete in price (discouraging developers from using said app stores), or for these alternatives to actually make a profit.
In short, I’m worried that Google will copy Apple, and find some nasty way to follow the letter of the ruling without really changing anything.
—
I am, however, still quite hopeful. As with Microsoft, change will come from outside. Before Google Chrome came, Internet Explorer was a broken, non-standards-compliant, insecure mess. Remember ActiveX? If you don’t, well, imagine having the occasional lick of poison in your water, and the water company saying “eh we need the poison in there to get you the water.”
There’s a test that’s designed to see how faithfully a browser complies with HTML and CSS standards called the ACID tests. The second version of this test, the ACID2 test, came out in 2005 — and Microsoft didn’t successfully pass it until 2009, with the release of Internet Explorer 8. (I’m not including pre-release versions here).
The third version, ACID3, came out in 2008, and Internet Explorer didn’t achieve a 100 percent pass until version 11 came out in 2013 — with Microsoft instead arguing that the testing conditions were excessive, arguing that scoring a 100 percent pass was “neither necessary, nor desirable.” Google, Mozilla, and Opera, I should note, passed very early on.
The point is, Internet Explorer was a dog, and it took actual competition — and user discontent — to bring about change. As much as I’ve complained about Chrome, it was a major step-up for users, and Google is only as dominant as no other company builds a better product.
And we’re already at that point. Competing browsers like Brave and DuckDuckGo are already offering better results than Google, and though I am loath to “hand it to them,” Bing is reliably better too. I think it’s very likely that Google’s ability to pay its way out of competition will have value, and while a negligible slap on the wrist is possible, it’s also far less likely with the amount of distaste for big tech’s massive revenues and terrible products that currently hangs in the air.
You may also be wondering what you, the reader, can actually do, and the answer is fairly straightforward: delete something, and use something else. While it’s hard to not use Chrome — I have specific plugins I use for my work — or Gmail, it is fairly easy to switch search engine, and even easier to find an old account you don’t use and delete it.
I know it’s impractical to switch your whole digital life, and that’s not what I’m asking. I’m just saying you should delete one of your accounts, and set up another somewhere else. Take the first steps towards lessening your dependence on these companies. A few years ago, CNET wrote a great guide about deleting all the data Google has on you, and you should follow it. You can go to Facebook right now and hit “delete past activity.” If you’ve got some old Microsoft account or Google account or Facebook account that you barely use, now’s the time to delete it. Give them as little as humanly possible, and be extremely intentional with the data you share. Consider moving to another browser, like Arc, or Firefox, or switching your search engine to literally anything other than Google Search.
These companies aren’t scared to lose you as a customer, but they should be. Every bit of information you share with them, every single service you use of theirs, every moment of your day you give them — and I recognize it’s unavoidable, I personally cannot leave Google Docs or Gmail without a significant upheaval to my workflow — is a way that they reinforce their monopolies.
So start deleting shit. Do it now. Think deeply about what it is you really need — be it the accounts you have and the services you need — and take action.
They’re not scared of you, and they should be.
