Why Every Platform Betrays You

problem from Facebook's perspective is that it's very brittle. Like the difference between I hate the service, but I can't seem to stop using it. And I hate the service and I'm not coming back is razor thin. So you get like a live stream mass shooting on Facebook and a bunch of users leave. And then investors get cold feet and they downrank the shares and share price tanks. And Facebook executives panic. But being technical people, they call it pivoting. And so one day, Mark Zuckerberg arises from his sarcophagus and says, you know, hearken to me, brothers and sisters, for I've had a vision in the night. The actual future is for me to transform you and everyone you love into a legless, sexless, low polygon, heavily surveilled cartoon character so that I can imprison you in a virtual world I ripped off from a 25 year old satirical dystopian cyberpunk novel that I call the metaverse. And that is truly the end of Unshittification, the giant pile of shit. Corey Doctorow, welcome. Thank you so much. Thank you. It's my pleasure to be on. I appreciate it. I know you're in the midst of what must feel like a nonstop book tour.

And congratulations on inshittification. And unlike some of your primetime interviews, you may inshittify to your heart's content. Yeah, there's no FCC for podcasts yet. No, no, that's the beauty. Not yet, at least. And I think, am I correct in saying this is book 30-ish? Is that— It's about 30, yeah. I lost count. And then there's like, there's essay collections and short story collections, and there's like a chapbook. So it depends on how you count. But it's a little over 30 books. And you got four or five more in the hopper from the looks of. Yeah. So there's two graphic novels, one of Unshittification and one of Unauthorized Bread. There's a book about how to be a better AI critic that's coming out in June. Nice. And there's a book that I've just pitched called The Post-American Internet that I hope to have out by like next winter. Wow. Can I get a spoiler on that? Yeah, it's basically I've been writing about it on my blog. So there's never spoilers. It's all just stuff that comes out of my blog.

But it's about the idea that much of the worst stuff online is the result of the U.S. insisting that other countries not, like, jailbreak and modify American technology as a condition of trading without tariffs. And so now that there's tariffs, that's not really a good reason to do that. And you could make a lot of money disensitifying American technology. And to make that case even stronger, Trump has now started ordering tech companies to switch off the administrative capacity of foreign nations he doesn't like. So this is Microsoft terminating the Outlook accounts for the International Criminal Court when Trump didn't like the fact that they'd sworn out a genocide arrest warrant for Benjamin Netanyahu. And so now you have governments all over the world that are being pressured by three different groups of people. So there's digital rights advocates like me. There's people who just want to make technology and make money by selling jailbreaks for iPhones.

So you can use third-party app stores or jailbreaks for printers. So you can use generic ink or what have you. And then there's national security hawks who are like, wait, if we can't jailbreak things, we can't replace the firmware for them. And, you know, maybe Trump will order John Deere to immobilize all of our tractors during the harvest season if we piss it off. Interesting. So I think that's a powerful coalition. Absolutely. I think it's a coalition that'll get stuff done. And I think it's not surprisingly an interesting conundrum that you draw out in that. And speaking of conundrums and speaking of inshittification, let's jump in. For those who may not yet be familiar, you coined the term to describe not just things getting worse, but a specific process. Walk us through the three stages, perhaps with one of your case studies. What are the actual mechanisms by which a company lands in inshittification? Sure. I like to use Facebook. I think they're kind of a prototype case here. Easy target. In stage one of inshittification, a platform is good to its end users while finding a way to lock those end users in.

So what Facebook did was in 2006, when they opened up to the general public and not just American college kids, they said, look, I know all of you already have a social media account on a service called MySpace, but I don't know if you realize this. MySpace is owned by an evil, crapulent, senescent Australian immortal vampire billionaire named Rupert Murdoch who spies on you all the time. And the last thing anyone wants is to use social media that's owned by a billionaire who spies on you. Come use Facebook. We'll never spy on you. We'll just show you the things you asked to see. And so we pile into Facebook and we get locked in and the lock in happens automatically with Facebook. The other services have to do pretty baroque things like, you know, Apple has to use digital rights management and Amazon has to convince you to buy your shipping in advance. So you'd be a sucker to buy your goods anywhere else because you've already paid for the shipping and so on. But when it comes to social media, something called the collective action problem, which is just an economist way of saying, you know, you love your friends, but they're a pain in the ass and you can't agree on like what board game you're going to play this weekend, much less when it's time to leave Facebook, means that you get stuck on Facebook.

You hold your friends hostage, your friends hold you hostage. So that's the end of stage one. And it brings us to stage two, which is exploiting the fact that users will now find it difficult to leave to make it worse for those users in order to make things good for business customers. So Facebook goes to advertisers and says, actually, we were spying on these people all along. Give us a small amount of money and we will target ads to them with incredible fidelity. And because we're such committed craftsmen, we're going to staff up this whole building full of engineers that are just going to police ad fraud. So if you give us a dollar to show an ad to a specific kind of user, we're going to find those users and just stuff that ad in their face. and they go to the publishers and they say, hey, you remember we told these rubes that we were only going to show them things they asked to see. That was a lie too. Put excerpts from your website on Facebook with the link back to your website and people will come and they'll click on the link. We'll stick it in people's eyeballs, people who never asked to see it. And that's your free traffic funnel. So the advertisers and the publishers pile in

and they get locked in really easily because when you're a business, even a relatively small amount of your customers can be absolutely vital. So if there's a single source of, say, 20% of your gross revenue, it's really hard for you to survive if that gets cut off overnight. You know, if you run a coffee shop and the building next door is an office tower that has customers for you that amount to 20% of your gross revenues and that business goes under and all those people go away and you see a 20% overnight drop in sales, that's probably it for your coffee shop, right? You're not going to be able to make the payment on loans. You're not going to be able to make payroll. You're not going to be able to make rent. So businesses are extremely sensitive to these supplier relationships. It's something that economists call monopsony. So we all know about monopoly because we've all played that board game on rainy days at the cottage until you wanted to kill your family and yourself. But monopsony is when you have powerful buyers, right? It's the inverse of monopoly. And every small business is at risk of being stuck in these monopsynistic relationships.

So once stage two is done and the business customers have been lured in and locked in, it's time for stage three, which is making things worse for everyone in order to harvest all the value for Facebook executives and shareholders. So ad targeting fidelity goes way down. Ad prices go way up. Ad fraud goes through the roof. In 2017, Procter & Gamble zeroed out its annual surveillance ad budget, which was like 200 million bucks a year for these surveillance ads. They saw a 0% drop in sales because to a first approximation, all those ads were just disappearing down the fraud hole. And publishers find that they have to put longer and longer excerpts on Facebook. Eventually, it's just your whole article. And you can't really even put a link back to your website because Facebook will treat that as a malicious link and downrank it. So now you're just like a commodity back end supplier to Facebook. You've replicated your website on their platform. And the only way for you to monetize that is using their shitty ad network. And then, you know, users find like all of the stuff in their feed that they want to see has been shrunk to a kind of homeopathic residue. And the void is being filled with things people are paying to show you that they're getting ripped off on.

And so that's the end of stage three, the giant pile of shit. And the problem from Facebook's perspective is that it's very brittle. Like the difference between I hate this service, but I can't seem to stop using it. And I hate this service and I'm not coming back is razor thin. So you get like a live stream mass shooting on Facebook and a bunch of users leave and then investors get cold feet and they downrank the shares and share price tanks and Facebook executives panic. But being technical people, they call it pivoting. And so one day Mark Zuckerberg arises from his sarcophagus and says, you know, hearken to me, brothers and sisters, for I've had a vision in the night. It turns out that while I told you that the future would consist of you arguing with your most racist uncle using this primitive text interface, I ginned up in my Harvard dorm room so I could non-consensually rate the fuckability of my fellow undergraduates. The actual future is for me to transform you and everyone you love into a legless, sexless, low-polygon, heavily surveilled cartoon character so that I can imprison you in a virtual world I ripped off from a 25-year-old satirical dystopian cyberpunk novel that I call the metaverse.

And that is truly the end of inshittification, the giant pile of shit. Man, when did you see this take form as this three-stage process? And, you know, I'm thinking of this meme, Corey, that floated around today, which was GAFA and Co., all the big tech CEOs gathered around a Cybertruck in a parking lot. I frankly still don't know if it was AI or not. but it's too plausible either way. And so my point being, it's tempting to say that this is collusion on a grand scale, but, you know, in your view, did it come into being as a almost lean startup kind of here are the five stages and the, you know, here's the loop we flow through? Like, when did this become concrete and obvious to you? So, you know, I think that in shitification, really is the outcome of unshackling a business from discipline.

And to explain that, let me say that the ideal form of a business, if you're an investor, you know, you mentioned lean startups and so on. The ideal form for an investor is a business that gets to charge infinity for substandard products that it pays its suppliers zero for. And while that is the business model of academic publishing, most other businesses can't get away with doing that. And so all businesses have to... The flywheel. Say again? The flywheel. The flywheel, yeah. So all other businesses have to accept some compromises. They have to share value with their stakeholders, whether those are end users or business customers or suppliers or other entities. And they have to take some measures to ensure that the product is of a certain level of quality and so on. And the things that impose that discipline on them, that force them to take those, you know, from their perspective, suboptimal takings from the value and the platform

are a mix of four factors when we're talking about tech companies. So there's competition and regulation, which are really two sides of the same coin. Because if you don't enforce competition law, if you let a company do what tech companies have done, which is buy all their rivals, You know, Apple buys 90 startups a year. You know, Tim Cook brings home a new company for its shareholders more often than you're bringing home groceries for your kids. Every successful product Google has had except for search is an acquisition. All of Google's internal projects just crash and burn. They suck at making things. They just operationalize other people's stuff. You know, Facebook bought Instagram when Instagram was taking Facebook users at an ever-growing rate. And Mark Zuckerberg sent a memo to his CFO explaining why he was doing it. He said, people don't like Facebook. They like Instagram. They leave Facebook for Instagram. They don't come back. When we buy Instagram, we can recapture those users. It's one of the many criminal intentions that Mark Zuckerberg has committed to writing an email to someone else, a thing that he does with alarming regularity, including things like it is better to buy than to compete, which is like if you're going to do a murder, you might as well premeditate it.

Right. So, you know, we let these companies buy all their rivals and then they capture their regulators, because when there's like five companies in a sector, it's really easy for them to agree on what they're going to demand of the regulator. They don't have people who defect from the consensus. When the regulator asks for public comment, all of the comments from industry read like they were written by the same people because they basically were. Not always because these companies are secretly meeting behind closed doors to set this up. Sometimes just because, you know, the only way you can advance through the ranks of a highly concentrated sector is to hop between firms. That if you're a VP and there's no SVP above you on the org chart and you've been there for a while, you go to one of the rivals. They just don't know each other. You know, Sheryl Sandberg was at all of them, you know. So they just get these tacit forms of collusion, of conspiracy. You know, when everyone who is a decision maker in a sector is either in a polycule or is the godparents of or is the, you know, to the kids of or is the executor of the estate of or is on the pickleball team with every other decision maker in the sector, then they just naturally come out to these conspiratorial outcomes.

And of course, they are just a slosh in money because when you divide up the market like the Pope dividing up the New World instead of competing, you don't erode one another's margins. So you have Google like bribing Apple to the tune of over $20 billion a year not to enter the search market. And this means that they don't have to engage in what Peter Thiel calls wasteful competition, which would cause them to, for example, offer advertisers lower rates, spend more money on quality, share more of the ad revenue with publishers, have more preferential terms for key publishers who are indexed by the platform. You know, all that stuff just goes away when you've only got one 90% market share search giant, which is Google. And so you lose the discipline of regulation, you lose the discipline of competition. And then there's two other sources of discipline. So one is the discipline of competition with companies that don't exist yet, of new companies being founded or new entities being founded to compete with you because you're doing something gross and terrible.

So, you know, we are blessed in the digital world with a very flexible kind of substrate. The only computer we know how to make is something called the Turing Complete Universal von Neumann machine, which is an engine capable of calculating every valid program. All computers can run all programs, which means that if your computer, you know, has a program that you rely on and the people who make it decide to install a 10-foot pile of shit in it, someone can make you an 11-foot ladder out of code that you can go straight over it. If, you know, if your phone only uses Apple's App Store or Google's App Store, someone can offer you the code to unlock the phone and take a third-party App Store. And if Apple takes away, you know, IceBlock, which is an app they killed last month that allows you to know when there's masked thugs in your neighborhood who are seeking to kidnap you and send you to a camp, you know, you can just switch App Stores and go to one that does that. But for 20 years, we've expanded IP law to make reverse engineering illegal so that this kind of complementary good has become effectively illegally or legally impossible to make, even if it's technically possible to make.

Right, right. So, you know, if you want to, like, jailbreak a printer so it takes generic ink, that's illegal because it's illegal to go around what's called an access control. Even if you're doing it for the totally lawful purpose of just, you know, putting new ink, you know, generic ink in your printer, the fact that this is the case combined with the fact that companies can buy their rivals means that the inkjet market has dwindled to about four companies that control the market. They are all colluding either explicitly or tacitly to raise the price of ink. Ink is now the most expensive fluid you can buy as a civilian without getting a special permit. At $10,000 a gallon, it would be cheaper to print your grocery list with the semen of a Kentucky Derby winning stallion. Right. And that's someone else's business. You know, when Jeff Bezos entered the market, he said to the publishers infamously, your margin is my opportunity. That margin on it, whose cost of goods is a penny a gallon. That's a big opportunity, but it's illegal to take it. So we lost that source of discipline, too. Companies don't have to worry about their regulators. They don't have to worry about their existing competitors. They don't have to worry about new competitors. But there was one more source of discipline in the market, and that was tech workers, because they were very valuable.

They're adding, you know, on average more than a million dollars a year to their boss's bottom line. And they're very scarce, which means that if your boss orders you to inshittify something that you care about, that, you know, you missed your mother's funeral to deliver on time, that you built because, you know, your life was transformed by technology and it helped you figure out who you were and, you know, break free of the destiny that was determined for you by where you were born or the milieu you grew up with. And you want to give everyone else those benefits. And you say to your boss, no, I refuse to inshittify this. If your boss fires you he can replace you And you can get another job at the shop across the street by lunchtime And so now that we have had mass tech layoffs and because tech workers didn consolidate the power they had from scarcity through solidarity by unionizing, because they mistakenly thought they weren't workers, they thought they were temporarily embarrassed founders. Well, you know, half a million tech layoffs later, tech workers do not tell their bosses to go to hell. And so we've lost

all the things that stop companies from shooting for that platonic ideal of infinite dollars, zero payment payout for suppliers, and, you know, the rest of us have to eat shit. Hard to argue. I mean, I think you and I both are framework enjoyers. Yeah, I love my framework. It's what we're doing this on right now, running Linux and et cetera, et cetera. You know, graphene running on my Pixel, all that. But it only takes us so far. And I appreciate that. and we'll get into that. I'd love to go in a little deep, Corey, on banking identity and speech, which to me are very, very blurry and slushy. And so you've talked about, I found one particularly interesting concept, half computerization. The idea that the problem is not that everything is computerized, it's that it operates one way. Vendors can update your device, change terms, revoke access remotely. You cannot do anything back. Help us understand how that asymmetry

enables the abuse of users and customers that you've drawn attention to. So regulatory capture has two sides. One is preventing the state from regulating you. And the other is demanding that the state regulate other people who threaten your sweetheart deal. And, you know, when we talk about inshittification, you see both of those things at work. So the expansion of IP law, the prohibition of reverse engineering, the cloud of rules that can stand in the way of scraping, of building alt clients, of even doing things like using protocol analyzers, much less capitalizing a business that jailbreaks something and makes it work in the way that, you know, the users might want it to work, even if the OEM, the manufacturer doesn't. That gives these companies the infinite flexibility to just change the rules at will. I call it the Darth Vader MBA. I have altered the deal. Pray I don't alter it further. So, for example, Qualcomm bought Arduino a couple of years ago.

This is this open source hardware microcontroller. It's really the gold standard of it. It is to open source hardware what the Intel 8086 was to personal computing or what Chrome is to browsers, right? It is just the thing everybody uses. And Qualcomm just changed the rules. They now reserve the right to assert patents against you. So if they don't like what you're doing, they can just shut you down. They require you to allow your code to be used to train AI. They've just like unilaterally altered all the rules. And you see this all the time. You know, Adobe switched to Creative Cloud instead of delivering you software on floppies or CDs or as a download. And then they could just take features away. The theory was, oh, no, we're going to use this to make sure you're always up to date. You're not going to have to, you know, download an 800 gig patch and then spend, you know, three hours applying it when you want to get some work done for a client.

But it also means that you fire up Creative Cloud one day. And like this really happened, Adobe decided to stop paying for what are called Pantone colors. This is a proprietary system for defining colors that you need a special ink for. so that might be a very bright color or color with metal in it or a color that has retroreflection in it or a color that is like conductive or that has gold foil in it or whatever if you're a designer you've got a whole you've got a thing called a pantone book which looks like the chips you get at the paint store to show you all the different paints it's this thick book you have a client come in and they're like okay we're going to do the poster and we want the writing and gold and you like go through the book and you find the gold they like and and then in photoshop or illustrator that just gets rendered as like an approximation of the color. But when it goes to the print bureau, it's literally like it tells them to go and like get a vat of that color gold and like screw it onto the printer so that it can be sprayed onto the paper as one of the passes when the paper goes through the printer.

You know, Pantone's a company. This is its product. It charges a license fee. And one day Adobe just said, no, we're not paying that anymore on behalf of our customers. And they went to the customers and said, hey, now that's 21 bucks a month on top of your Adobe license. And if you don't pay it, we're just going to turn all the Pantone colors in every image you've ever made in your entire career black. So we're just going to take the color out of this. So this is like the unilateral power of an infinitely flexible digital computer that is in the control of the manufacturer. And then on your side, if you want to change it back, right, if you want to get a plugin or an alt client or anything else that would just like put those colors back or put those features back or restore privacy to a product that used to give you privacy and now doesn't, you just aren't allowed to reverse engineer it. You know, there's one of these IP laws that I focus on a lot in my work is something called anti-circumvention law. So this refers to a law that bans you from modifying

a technology if the manufacturer has in some way said you can't modify it. So in other words, if like in the software code for the device, if the programmer has done the equivalent of like drawing a chalk outline around some of the code and scribbled next to it, do not look at or modify this code, it becomes under Section 1201 of the Digital Millennium Copyright Act, a felony punishable by a five-year prison sentence and a $500,000 fine to change that code. It's what Jay Freeman calls felony contempt a business model. Because although Congress and Parliament have never sat down and said, you know, you're not allowed to install an ad blocker for an app, or you're not allowed to use generic ink with your printer, or you're not allowed to update your smart speaker so that it does all the interesting stuff locally instead of feeding continuous telemetry back to the manufacturer, or you're not allowed to replace the firmware on a device that the manufacturers decided to stop supporting. In the book, I talk about people who had these bionic eyes, people who had

been born without eyes, who had severe nerve damage in their eyes, or blind for other reasons, who were able to wire effectively a camera into their optic nerve permanently. And then the manufacturer went under and they decided that they didn't want to release the source code so that other people could support it. So these people's eyes just stopped working, right? And so if you want to like make new software for that bionic eye, the reverse engineering step is illegal. And so you and I are defenseless, right? The law will step in to destroy us if we modify our own property to work the way that we want it to. But the law will step in to defend the manufacturer if they modify our property so that it doesn't work the way that we want it to. And so this is This is both facets of regulatory capture. It is both existing without regulation and mobilizing regulation against people who do things you don't. If we look at payments, I know you've written, Corey, about Operation Choke Point 2.0.

You have laid out your case certainly in the book and hear about why platforms are in a position to betray users. And thirdly, I know you have strong feelings about Bitcoin, but let me use it as an example. So the point being, where do you come down on protocols over platforms, right? So for those who are building on Nostra, building on Bitcoin, building on Keat, you pick it, whichever one you find most attractive. Where, in your view, is the fallacy or the folly in saying let's just bypass the platforms and head straight for the permissionless protocols, understanding that it's harder, understanding that monetization and all these things become real challenges? What's your perspective on that protocol versus platform debate? Well, I just don't think money is a platform. So I think that saying, well, we've now made money into a protocol and not a platform presupposes that money is a platform.

Money arises out of – it's a thing that states create and that the thing that gives it liquidity and value is that states tax it. That's the origin of money. There was no evidence whatsoever that coin money spontaneously evolved from people who couldn't, you know, figure out how to make change for two goats and chickens. And so they spontaneously arrived at gold. And then states came and took their gold away. The origin of money and the source of money historically and the source of money in the world today is taxation. You have a state that wishes to provision itself. So it lays on a tax. It says, everyone here owes me some taxes. And then it goes out and says, hey, the coin that you owe taxes in, I will give you if you perform services for the state. And that is what creates liquidity. So without central banking, right, without something deciding on the monetary supply, if you peg the monetary supply to something like gold, you just get this incredible instability.

It makes for the shittiest money imaginable, which is why you got a series of panics and bank runs and which is why Bitcoin sucks as money. Right. It's not a store of value because the value yo-yo is around. And this is even more true of shit coins. Right. If if if Elon Musk can double the price of, you know, double the amount of money you owe on your mortgage or cut it in half based on tweeting like a doge and a poop emoji, then this is extremely bad money. It's also a terrible unit of exchange, first of all, because of the transaction costs, but second of all, because of the volatility. It's a terrible unit of account for the same reasons, which is why the only thing you can buy with Bitcoins is shitty monkey JPEGs and other Bitcoins, right? Like, what is for sale in Bitcoins? Well, and I mean, if we had another two hours, and let me just say genuinely, I would love for you and Safety and Abus to have a debate. But my point, honestly, is not to debate Bitcoin. It is in pick a protocol, right? So, for example, I know you use the Fediverse Mastodon.

Sure. Which, you know, itself is just a bunch of servers, right? And so my point is, do you see a protocol or a truly permissionless protocol, do you see that as a valid redress approach to some of these platforms and their inherent abuses? Well, so it's, look, like building a thing that is more unshittification resistant because it is harder for firms to, first of all, merge to monopoly and therefore do regulatory capture. Second of all, block third parties through shutting down APIs or blocking interoperability because it's based on open protocols and open source and free software and so on. All those things are great, but it's not a solution in and of itself because people are trapped on the platforms by the collective action problem and switching costs. And it's true that if we just like leave the platforms to themselves, eventually they might implode because they'll just get so terrible that people are willing to endure the switching costs.

But that's not a thing we should want. I mean, the reason people are on Facebook now is because they love their friends more than they hate Mark Zuckerberg, and they don't know how to reestablish their community elsewhere. You know, so there's a friend and colleague of mine, really a hero of mine named Andrea Downing, who runs a nonprofit called The Light Collective. And Andrea is a breast cancer previvor. So she has the breast cancer genes, and she was part of an online community for breast cancer previbors that was lured in by Facebook in the early 2010s when Facebook was courting medical communities. And this is a really important community, the breast cancer previbor community, because if you have this gene, first of all, you are constantly thinking about these incredibly consequential medical decisions you need to make. So like, should you have your ovaries, uterus and breasts removed? Right. These are hard decisions to navigate. But you're also, if you carry this gene, related to a bunch of women who are sick or dying. Right. You know, your daughters, your mothers, your, you know, your wives, your grandparents, your grandmothers and so on are, you know, also carrying this gene and also sick or dying. And so this community is super important.

And Andrea has been trying to get this community off of Facebook for quite some time, ever since she just accidentally, she's not a really technical person, but she just discovered that you could enumerate the full membership of any Facebook group, even if you weren't a member, which is incredibly damaging to a medical group like this. And so she reported to Facebook and they won't fix it. They said, no, no, no, that's not a bug. It's a feature. It's part of our ad tech stack and we're not going to change it. And so they actually sued Facebook over this. But when the FTC settled all outstanding privacy claims against Facebook as part of that multibillion dollar settlement, that lawsuit died and can't be reopened. And they're still on Facebook. And it is possible that if we just leave Facebook to its own, Facebook will become so horrible that they're like, the value I get out of having this help to navigate these medical decisions, having support for these terrible tragedies underway in my life. All of that is worth so much to me, but Facebook is so painful I'm going to leave.

And then we have the worst of all worlds because they've lost their support networks and they've endured this Facebook pain, right? What we want to do is not shatter the platforms. We want to evacuate them. And so, you know, right now— Help me understand. I'm not sure I can discern. And I'll just interject and say I know you spent, you know, decades talking about the right to leave, interoperability, data portability. So map that out. What does success look like in that? When Facebook was struggling with luring users away from MySpace, it wasn't enough to just pitch them on having a superior privacy policy. Because, like, no one realistically was going to leave MySpace and hang out all alone on Facebook, rereading the privacy policy and waiting for their stupid friends to wise up and join them there. And so Facebook gave those users a bot and you gave that bot your login and password and it would go to MySpace several times a day, scrape everything waiting for you in your feed and put it in your Facebook feed, which you could reply to and push it back out again. So if you did that to Facebook today, they would nuke you until you gloat, right? They'd say that by reverse engineering the app, you violated Section 1201 of the Digital Millennium Copyright Act. By violating the terms of service, you are both a tortious interference, interferer with their contracts, but also that you violated criminally the Computer Fraud and Abuse Act by exceeding your authorization on their servers.

that anyone who was ex-Facebook who helped you was violating their trade secrets, that, you know, they just like that you were violating their copyrights, their patents, their trademarks, right? They just like destroy you, right? So like on the one hand, we can make a way for people who've left Facebook to continue to receive messages from Facebook and send messages to Facebook and kind of allow their friends to trickle off of Facebook rather than having to all agree and make the leap. And it's much easier to go when everyone you know is now somewhere else, right? If everyone, you know, especially if you like put a little footer in every message that you send that says this message sent from Mastodon, right? And so it's much easier for people to make their way over. But also we could mandate that Facebook support a protocol, right? So like both at proto and Mastodon and activity pub more broadly have ways for people to leave one server to go to another and to have their messages forwarded, to have the pointers move over.

So we could we could regulate Facebook and say you must support at proto or Mastodon. Or we could say, you know, to get really in the weeds here. What I think we should do is we should say these are the technical functions for an interoperability protocol that we want you to support. You can roll your own. And if you are found to be out of compliance, we're going to hit you with gigantic fines. And maybe if we think it's willful criminal prosecution, if you want to be in a safe harbor where you are presumptively in compliance with the rule, then you then you implement either at proto or the master on protocol. And so question. Yeah. When when and I hear you, I'm I'm I'm monitoring my internal pessimism. When has that worked and why should we believe that they wouldn't run it or undermine it? It's worked for the most notorious telecoms monopolist in the history of the world. The bell system. Fair, fair.

Right. And, you know, if you want to switch from T-Mobile to, you know, Google Fi, you do 10 seconds of admin work and your number just pops over. No one cares which network you're on. If you want to like it, no one's ever phoned a friend and said, dude, you wouldn't believe whose SIM is in my phone today. Right. And so like it works. And these are rapacious convicted monopolists. Right. Now, do you think, I mean, so here's my follow-up, which is fair. I mean, I recently switched to CAPE. Sure. I'm not paid by CAPE. All those VRMOs, all those virtual network VNMOs, the virtual network mobile operators, they exist because of number portability. Absolutely. Now, question, is it, do you think, was that just, you know, what, 150 years of trying? And do we need to expect another 150 years before we have the ability to port out of Facebook? I mean what gives you sort of hope or reason to believe that it could and would happen on a more expeditious timeline So I agree that it very hard to regulate companies once they monopolists right So the historical standard for anti was something called anti

So you would stop companies from attaining a monopoly on the grounds that once they had the monopoly, it's very hard to boss them around. And you said 150 years with AT&T. It wasn't 150 years. It was 69 years to break up AT&T. Okay. Well, you know, I was being facetious. But it was a long time. And, you know, the challenges, the barriers to breaking up AT&T look a lot like the barriers to breaking up tech giants today. So in the mid 50s, AT&T came within a whisker being broken up. But the Pentagon intervened and they said, we cannot win foreign wars. And in particular, we will lose the Korean War if you break up AT&T. Now, you may know that they did win. They did lose the Korean War anyway. But AT&T won the Korean War because they got another 30 years out of it where they could continue to extract monopolies from us. And it was really hard. It took quite a long time. You know, and it's not just AT&T. IBM was almost broken up. And the way that they avoided it was so they were sued by the DOJ in 1970, the DOJ Antitrust Division.

And that case lasted until 1982. And every year for the next 12 consecutive years, in a case that was called Antitrusts Vietnam, IBM spent more on lawyers to fight the DOJ lawyers who were fighting IBM than the DOJ spent on all the lawyers fighting every antitrust case in America. IBM was able to outspend the U.S. government for 12 consecutive years, and they ran out the clock. Reagan got elected and he dropped the case against them. Actually, what happened was that Reagan's advisors were convinced that AT&T and IBM were together strangling the American tech industry, and this is why tech had stagnated. And they said, you've got to get rid of one of them, but you're going to look like an asshole if you do because you just ran on like this deregulatory platform. So we think you should go after AT&T because you can claim that they weren't really a for-profit company, that they were a regulated monopoly. So you're basically just breaking – you're doing a doge, right? You're like, these guys actually are the government, and you're just shutting them down, right?

You're privatizing a government agency effectively. But, you know, that was how we got there. So it's a long goddamn slog. It's really tough. But you remain hopeful. That's a genuine question. So there's two sides to this, right? One is the adversarial interoperability, which is when you're jailbreaking them and modding them and all clients on device bridging. The other one is mandates. And it's really easy to cheat on a mandate, right? Like there's a lot of room to like say, oh, you know, we shut down this gateway because we thought we'd been hacked and we were hemorrhaging user data. And the fact that we do this every like, you know, 30 minutes and it makes it impossible for anyone to like have the interoperability. That's just an unfortunate epiphenomenon of the difficult cybersecurity environment we exist in today, right? It's really hard to do. And like figuring out whether they're lying is fact intensive because, you know, everyone who understands Facebook's infra is a Facebook engineer. So how do you even figure out if their technical claims are correct? You know, so this is a really tough thing. But adversarial interoperability, this reverse engineering hacking, all client stuff, poses an absolutely unquantifiable risk to tech firms.

Because you're basically saying, whereas today, if someone starts reverse engineering your stuff, right, and you're Mark Zuckerberg, you walk out of your office and to your right is a building full of lawyers and to your left is a building full of engineers. And the engineers generate profit and the lawyers generate costs. So you walk into the lawyer's office and you say, send these guys a letter and terrorize them. And if they don't respond to it, ruin them, salt the earth and make sure no one ever invests in a company like theirs again. And that means that the people making you money can still do their job. Right. But if the lawyers say to you, I'm sorry, the law no longer allows us to mobilize the state to fill in for our engineering staff, then you have to task an unquantifiable number of your engineers for an unquantifiable amount of time to fight a holding action where the other side has the attacker's advantage in that they only have to find one bug in how your code works and exploit it.

And you have to have the defender's disadvantage, which is making no mistakes. And moreover, they can find 50 bugs and they can operationalize one of them. And the minute you shut that down, they can get it. Yeah. So now you have this different equilibrium where if you want to stop firms from doing this reverse engineering stuff that you hate even more than you hate having fair competition in the market, the only way to do that is to make an API that's so robust that you'd be an idiot to just try and reverse engineer. of the platform because you've got like a documented API that works, right? So now you're setting up a totally different set of incentives. Now, not all firms are going to follow it because tech leaders are extremely hubristic and stupid. But when they do the stupid thing, then you can fall back on the API. And so what you get is a thing that works a little like a two-part epoxy, where you have a mandate that is quite brittle, but very strong, right? If the API works, it works. So you don't have to change your code every two days as they patch the bugs.

But then you have something that's very malleable, but doesn't hold things together very well, which is these reverse engineering solutions. And you put them together and you get something that's supple but strong, right? You don't have that brittleness anymore. So that, I think, is our path to glory. And how does that is – I know we could spend a great deal of time here. What does that look like in practice? Is that at the state level someone takes the lead? Is there an example that comes to mind? Is that sweeping federal regulation? How does that get done? I think it's got to be feds. And I think because some of these laws, like all the copyright laws are federal with preemption. So they're like all state copyright laws were preempted in the, I believe it was the 1976 Act. And you cannot make a state copyright law today. So you can't make a state exemption from a copyright law. Now, there are ways around this. So Oregon's Electronics Right to Repair Law and the pending Washington State Right to Repair Law both contain prohibition on using DRM in a way that interferes with repair.

So like if you've put DRM on the digitizer assembly in an iPhone, so the screen and its underlying electronics, so that when a repair person swaps a new digitizer in, it won't be recognized by the CPU until an Apple authorized technician adds an unlock code. So they can't immunize people from liability who reverse engineer that unlock code. Well, what they can do, Oregon as a sovereign in charge of in-state commerce, is they can say it's illegal to sell a phone that works that way in Oregon. A phone that works that way is defective under Oregon law and you cannot sell it, which is what they've done. And so this applies to John Deere tractors and ventilators and phones and all kinds of things, cars. So that's one way the states can get around it. It's probably going to have to be federal and it probably won't be American. So like I was saying before, this post-American Internet, I think this is a thing that is going to define the coming years of computing because every country in the world has engineers. Every country in the world has capital.

And actually, these days, they have extra engineers and extra capital because there's a lot of engineers fleeing America. And there's a lot of capital that would like to be able to invest in an environment where the biggest predictor of success isn't how many Trump coins you buy. So there's capital and there's tech talent everywhere. And Trump is making the case for doing it. So the only reason other countries have anti-circumvention laws in their books is because the U.S. threatened them with tariffs if they didn't pass them. And now they have the tariffs. And look, if someone says, you know, you have to follow my orders or I'll burn your house down and you follow the orders and they burn your house down, you're a sucker if you keep following the orders. Right. And, you know, when I think about change, I always think about what the coalition looks like. So, you know, anytime you see a group of people who've been working to get something done for years without any success and then suddenly they start to succeed, it's because they found a coalition. Right. It's because it's not because they've like come up with a cool new hack. It's because they found a bunch of people who don't agree with them on everything, but agree with them enough to do this thing.

And so, you know, for example, Trump, right? Trump has figured out how to well together billionaires, bigots, you know, people who've got low tax brain worms and would vote for a slime mold if it would take a nickel off their taxes. You know, and then people who are like, fuck the Democrats. They're a bunch of do nothing, say anything losers who don't help me out and who don't give a shit about me, which I think is broadly true, at least of the party establishment. And so he's welded together this coalition of people who, you know, like mostly don't like each other. Right. But they're like willing to work together. And that's why every time there's like a little hiccup, all the news just consists of like the Trump coalition is shattering because like it might. Right. It's people who don't like each other. So, you know, what's the coalition look like for this? Well, I tell you, it's not enough to have the coalition of privacy advocates, labor advocates and consumer rights advocates who've been trying to get rid of this stuff for 20 years.

Because we have failed, right? And it's arguably not enough to add in the people who would like to make hundreds of billions of dollars raiding the margins of the most profitable companies that have ever existed, because they've had since Liberation Day. And so far, it's not on the agenda. But I think when you throw in the national security hawks, who aren't just national security hawks worried about Trump bricking their tractors, but who are like legitimately worried about the Chinese bricking the solar inverters in Europe or the batteries. What Trump is doing is the stuff we were told Huawei would do if we let them into our telecoms infrastructure, right? And he's actually doing it. And it wasn't unreasonable to worry about the Chinese doing it, not least because when you have a backdoor in your or, you know, and a backdoor can just be an over the air update mechanism to which you can push bricking code. Right. When you have the backdoor mechanism in the device, it can be used by the manufacturer. Right. It can be used by a government who can order the manufacturer around. It can be used by an insider working without the manufacturer's knowledge, but who has access to the manufacturer's credentials.

And it can be used by a criminal who impersonates the manufacturer. And so we are now at this center. And in fact, almost certainly by all of them together. By all of the above. Yeah, absolutely. And this is the story of CALEA, which is the backdoors in switches sold in America. So you have this incredible coalition that is kind of being born now. It's quite overdue, in fact, when you think about it. It's really silly that we have all this stuff organized the way it is. And it's part of a trend that Abraham Newman and Henry Farrell have written about in their book, The Underground Empire, of American infrastructure that was being used globally as a kind of neutral trading ground or, you know, kind of a lingua franca being operationalized or instrumentalized or weaponized by the Americans and everyone else having to figure out what to do. So it started with like Snowden and Snowden saying, you know, it makes it turns out to be a giant mistake to have all the fiber in the world terminate either on the Atlantic or Pacific coast of the American landmass.

Right. So now everyone's out there. Yeah. And so now everyone's out there playing like order and squared running point to point fiber between all the countries. This is a very difficult and expensive thing. And then there's the weaponization of SWIFT. Right. And and so that started not with the sanctions against Russian oligarchs, but actually the weaponization of the dollar starts with the vulture capitalists who sued Argentina over default on their bonds. And then they won the right to walk into the Federal Reserve and take all of Argentina's foreign reserves. And so then you got the Russian sanctions and SWIFT blocking. And now everyone in the world is trying to figure out how to do pairwise transactions among every currency in the world without using dollars as an intermediary, which is another order and square problem and very hard to solve. But when we talk about, like, liberating ourselves from American infrastructure, it's actually the only one of these that's easy. Because when you say, okay, well, we're going to replace all of the firmware and all of the administrative software with open source, auditable alternatives that can run on commodity hardware on metal and data centers in your own country, then you give rise to a kind of digital sovereignty grounded in international nationalism.

We're sure like Europe could put the seed funding into like replicate Office 365, but the Canadians and the Brazilians and the, you know, Ghanaians and the Russians could all be contributing patches to it. So is it a new kind of flag theory? I mean, is that kind of what you're pointing to? It's a kind of technology that works more like science. So, you know, in the same way that like even during World War II, there wasn't such a thing as Nazi radio technology as differentiated from allied radio technology. It's just radio, right? It's just like it's a science, right? Unlike, say, you know, the advanced GPUs, right, where we really do have like different we have big differences in terms of national capacity to manufacture these things. You know, like anyone who can like read a library book can get up to date on the most advanced physics.

Right. Well, not anyone. Anyone who has the proclivities for it can get up to date on advanced physics. And even nations that are enemies, their scientists are still publishing the same journals and going to the same conferences and so on. So there's a great deal of internationalism at the science level. And the way that we've practiced technology has been pretty unscientific. scientific, right? If you think of the foundation of science as being the split from the alchemical method, and so an alchemist observes two phenomena in the universe, hypothesizes a causal relationship, designs an experiment, runs the experiment, but then doesn't tell anyone what they think they've learned, which is how you get alchemists for 500 years learning in the hardest way possible that it's a really bad idea to drink mercury. And it's only when they start publishing, right, when they start exposing their knowledge creation to adversarial peer review and the humiliation and pain of being, you know, debunked by your worst enemy who gloats as they find the stupid

errors that you've made, that we actually turn the base metal of superstition into the precious metal of knowledge. And, you know, we don't do that with technology. It is still the case that the majority of our technology is made by people who won't tell you how it works and increasingly is designed so that it's illegal to try and find out how it works. Or is made a state secret, I think. And that's, you know, I don't know if you noticed today. Well, there's some secret. There's not much state secret. Like Office 365 isn't a state secret. No, no, no, certainly. But we're saying the Office 365 apps is a felony. Right. And I think, I don't know if you noticed today, Corey, that the GrapheneOS team pulled up stakes from France. From France. Moved to Canada. Yeah. And so, I mean, how does that factor in? And I, you know, as you were enumerating this coalition, including, which I thought was really interesting national security hawks. What occurs to me is, OK, here's this countervailing force of countries like France, you know, first holding the Telegram founder captive for a year

plus and and now leaning pretty heavily into the Graphene OS team. Like, how does that sort of frame into the picture you're painting? So NADSEC hawks are some of them are really stupid and they embrace a doctrine called NOBUS, which stands for no one but us, as a no one but us is smart enough to figure out this method. And so we can ignore the idea that there is no security in obscurity and we can demand that a backdoor be put into something. But that's not a uniform view in NATSEC. So like NATSEC itself is a coalition and within the coalition, one of the major fracture lines is disclosure of cryptography and security techniques and tools versus secrecy and suppression. And, you know, the secrecy and suppression people are like the reason we have a ransomware epidemic, because it's ransomware is built around a bunch of vulns that were hoarded by the CIA and the NSA and that leaked.

Yeah. And that leaked. Right. So Eternal Blue being the big one. And so like one of the reasons that at the start of the ransomware epidemic, it was really, you know, grimly funny, that you would have people who are just obviously dum-dums, who are nevertheless landing these incredible ransomware scores, where like your hospital, you know, a major hospital in a major city would be bricked by a ransomware gang. And they would say, we will not unlock your hospital unless you give us $200. And people would be like, $200? Like, are you 12? And they were, right? They were just like dum-dums. Because, you know, Well, what happened was the NSA created or the CIA with with Eternal Blue created. No, it was the NSA created this like, you know, pluripotent eternal omni weapon, which they were just like, oh, yeah, we'll just like leave this here and no one will ever rediscover it. And no, and it'll never leak. And then it did. And then and then just the dumbest people on Earth were able to hijack whole goddamn hospitals. Yeah. Never attribute to malice or it can be given to stupidity.

Yeah So you know like NATSEC needs to pull its socks up clearly right And figure this stuff out And you know if we want to talk about the power of this new coalition of people who want to make a lot of money people who are user rights advocates or digital rights advocates and NATSEC people one of the things that the allies of the pro methods side of the NATSEC world in those other camps can do is shout down the fools who have dominated their procedures for decades and have made secret hoarding and security through obscurity the, you know, failed god of national security. What would that, what is, paint a picture of that particular instance where you're calling to someone to shout down, what's the room in which this happens? What's that scenario look like? So like, you know, imagine the European Commission

builds a bunch of tools, right? They build like an Office 365 clone, but there's not really a good path to migrate data to it. And it's incomplete. It's kind of like they built a bunch of housing for people in West Berlin, in East Berlin. I think Denmark has claimed that they're going to move wholly off Office 365, for example. So you build a bunch of housing for people in East Berlin and you build it in West Berlin and then you sit around wondering why no one's moving into it, right? And it's like, well, have you guys noticed this wall over here, right? We've got to get rid of it. And so now like in the commission, in the national governments and among technologists and entrepreneurs, you're having debates, right? And you have people who are like, Well, I've raised a bunch of money to build the circumvention tools to give you one click transition. And you have other people who are like, you know, we should be really worried about how the Americans and the Chinese are in our infra. And we should we should have that stuff all be open and free. And then you have other people are like, no, no, no, it can't be open and free. Because if we tell people how it works, they'll find the errors. Right.

And so now you've got these other stakeholders, right? You've got the consumer rights stakeholders. So think of, you know, Meredith Whitaker every time who runs Signal every time there's a chat control proposal who's like, you're stupid. Here's why we won't do it, right? So you've got the user rights advocates on the one hand, right? Then you've got these people who are like, I have raised hundreds of billions of euros in capital. I have technologists here who want to mobilize it to build products that will make a lot of money for European firms, not least because we're proposing to do things like replace the firmware where all the services revenue comes from on phones and tractors and whatnot without having to endure any of the capital expenditure associated with manufacturing them. So we're just going to make John Deere and Apple assume all the risk and we're going to cream off all of their profits. And we would like to become rich as fuck, please. Right. And then you have the people in the national security debate who are either on the stage or behind the scenes arguing about this. And they're shaking the briefings being made by rich people, by technologists, by user rights advocates and saying, like, how can you listen to the morons who gave us the ransomware epidemic?

Isn't it time that we, you know, like disqualified them, you know, from being able to intervene in these debates? They're unserious people, you know? Yeah. No, that's a helpful picture. Sure. If we shift, Corey, then to lobby us, the end user, you know, you have said famously, you can't, you know, voting with your wallet ain't it. You can't shop your way out of a monopoly, I think is your phrase. And so a couple of examples, the Para app for DoorDash drivers, the Tuyul, I'm probably mispronouncing that terribly. Tuyul is how you pronounce it. As far as I know, Tuyul is how you pronounce it. In Indonesia, I believe it is. And so brilliant hacks in effect, but they're still operating on top of the platforms that are in shitifying them. And so my question is, at what point does, you know, sort of resist from within? And you've addressed this, but at what point does it have to become just leave and build something else, notwithstanding what you have said prior? Well, you hollow it out, right? You don't just shatter it. You hollow it out.

But, you know, if you, if, you know, a group of people who are currently, like if breast cancer previvors, if a plurality of them end up in a distributed group that spans different kinds of federated social media, you know, maybe it's grounded in a private Mastodon server, but federates with other ones and, you know, they're private accounts and you have to follow them or whatever. And now there's like, it's like 70% of the group or even 40% of the group. And Facebook shuts down or Facebook, you know, reaches its investor crisis. It's easy for those people who are now leaving to know where to go, right? It still might be traumatic. Some of them still might get lost on the way. But like, I'm trying to think of a parallel, you know, when my family are Soviet war refugees. And like, by the time after the war that my grandparents got to Toronto,

there was a neighborhood full of Jews from Eastern Europe who were all war refugees, who'd all come over on various boats over several years. And they kind of all formed a mutual aid society, they formed a community there. It wasn't exactly, it's not an exact analogy, because, You know, my grandmother lost touch with her family for 15 years and, you know, was never able to completely reconnect with them because they stayed in Leningrad. But, you know, like that's the that's what you want, right? You want there to be like an obvious place you go. Like if you're Ethiopian and you're leaving Ethiopia, you go to D.C. or Seattle or a couple other places where there's a big community. You're likely to have a bunch of friends and cousins and people from the old country. and like you can shop a grocery store that has the same food that you like and so on. It's not like being the first Ethiopian to show up in wherever Minneapolis. Right, right. And I mean, are there, and I'm sure we could take a lot of time on this, but are there examples that give you hope that that is happening or at least could happen?

And that could be, you know, you've talked about Fediverse, Mastodon. I'm a big proponent of Nostra. Would love to see you there, by the way, Corey. Well, you can by following me on Mastodon. Who cares which SIM I have in my phone? Who cares which SIM I have in my phone? Yes, indeed. Yes, indeed. You know, what, in addition to these protocols, gives you, maybe hope's the wrong word, gives you evidence that that destination or those destinations are emerging, which allow the hollowing out? You know, where do you see bright spots? So for like 20 years, almost all technology policy, digital technology policy, notwithstanding telecoms, was grounded in the idea of like making the platforms behave themselves better. Right. Like don't, you know, like like either replace Mark Zuckerberg with a better Mark Zuckerberg or make Mark Zuckerberg pull up his socks and act like a decent chap. And and, you know, so we have content rules and we have, you know, boardroom fights and all kinds of things.

And these were all like always a dead letter. They were always dumb. They were always grounded in the idea that the problem was that the wrong person was the unelected social media czar for a life of four billion people and not that the job shouldn't exist at all. And over the last like five years, there has been just a proliferation of regulatory policies that like neither I nor anyone like me had anything to do with that were about interoperability. So the Digital Markets Act and the Access Act in the United States and so on, these kicked off not because like EFF wrote model legislation or whatever. They just are kind of in the air. This idea that interoperability and, you know, mandatory APIs and gateways are like part of the way we solve this problem. And, you know, they're bipartisan. They are international. You know, the Chinese Cyberspace Act has a prohibition on Chinese tech giants blocking interop from new market entrants. So like it's everywhere. It's like it's in, it's, you know, it's not just bipartisan. It's not just bicontinental. It's also like across different kinds of, like radically different kinds of political systems.

So there is like a leading edge of understanding that, you know, it is protocols, not products, that it is about decentralizing things, that we do need to do science and not alchemy, that, you know, the fact that a firm can make more money by not telling you how something works is no reason to tolerate that. Like if we said if someone said, well, we're not going to tell you how we calculated the load stresses for the joists in this apartment building, we would say, well, then fuck off. No one gets to live in your apartment building. We're tearing it down and hiring someone else to build it. Right, right. And that I mean, that is hopeful. I mean, I guess the steel man I would offer is to say, you know, perhaps you'll disagree is is that I hear you arguing for better antitrust enforcement, stronger labor power laws. Yeah. Yeah. Neurot mandates. But I would argue that your book shows in great detail that regulators get captured, labor got crushed, laws get written by monopolists. You know, even with the hopeful anecdotes and points that you've made there, why should we trust that the next round of regulation won't also get captured?

Well, because you're missing a causal step, right? Let's distinguish between two different kinds of regulation. There's a bad regulation. Can we stipulate that there's such a thing as a good regulation? Oh, certainly. Yeah. So like you like being able to turn on your tap and you wouldn't want like a caveat emptor, you know. Absolutely. Water treatment regime. Water treatment, yeah. Yeah. So we have to distinguish between two kinds of bad regulation. Regulation where someone made a mistake and regulation where someone induced a mistake, right? And like I'm not going to say that – Would you not – let me just interject. Would you not say that there is a third and perhaps dominant category of just sharing competence or design by committee? Well, that's just made a mistake. That's another way of saying made a mistake. Right. So like, look, you know, though, again, notwithstanding like Flint, Michigan and a few other places that are ghastly stains on our national character, like even though there are no water chemists in Congress, you can pretty much just turn on the tap and drink the water.

It's not because like clueless people don't understand technology and therefore we can't make rules. Like the way we got rules that let us drink our tap water was through a system of expert agencies that do notice and comment into a market that is pluralized. So you don't have people who induce mistakes by capturing the regulator or by overwhelming the regulatory process with junk science and junk submissions. And if you want to see what it looks like when that's not happening, look at the UK where they privatized the water system. And now you have these firms that have ganged up on their regulator. They're bigger than the regulator. They're more powerful than the regulator. And they had like 4 million raw sewage discharges into the navigable waters of the United Kingdom last year. Every navigable water system of water in the United Kingdom is now considered unsafe. When they had the Oxford Cambridge boat race last year, I think all or nearly all of the athletes got listeria from splashback from the Thames. Wow.

But those were induced mistakes. Those mistakes were induced by the system. Forced errors. Yeah. My point isn't that regulators will always be perfect, even under ideal conditions. It's that unless you have a system in which you have pluralized power that is distributed among many players so that you can have the kind of scientific method in which, you know, people of equal power are able to debunk one another's claims and to get at the truth in a truth-seeking exercise that we call regulation, then you will always have failure. So the only way to get a good regulation is to break things up. That's, again, why anti-monopoly law was so central to the project of the long period of American prosperity. Right. It's like that period is not just a period of economic growth because you have easy market entry and you have, you know, a real dynamism where dominant players can't suppress exciting new technologies that gore their ox.

But also a period of good regulation, broadly good regulation, because you also had the conditions under which experts working for the state could adjudicate claims by experts working for industry and experts in civil society to arrive at something that was a rough truth that was good enough to get things going. And, you know, like, I'm not saying that it's perfect. Even when it worked, it might not be perfect, right? Like, you know, if you insist that everyone overbuild their joists, then you are going to deprive people of housing because you're going to increase the cost of housing stock and you're going to, you know, inflict environmental damage by requiring more material extraction to build those joists and so on. And there's a virtue in getting that number right and not just overbuilding it. But, you know, like the way that you get there is, again, by having a process of honest regulation, right? It can't just be market forces because when we just allow market forces to run this at all, you end up with shit in the waterways, right?

Talk about in shitification, right? You end up with shit in the waterways. Yeah. Yeah, I mean, decentralization, more of it, please, I think across all. And I always tell my libertarian friends, you know, if you want just a government that enforces contracts and nothing else, they still have to be more powerful than the entities whose contracts they are enforcing. Otherwise, how are they going to enforce them? So the smallest government you can have is determined by the largest corporation you're willing to tolerate. And if you want a smaller government, have that government first and foremost enforce antitrust law. And block incipiency and not just try to address monopolies once they form. That's poignant. Absolutely poignant. As we wrap up, Corey, someone listening to this, you know, they're thinking, OK, Google controls my email. Apple owns my phone. My bank can freeze my account at their whim without explaining why. What's one or two concrete moves they should take next?

Well, step one, join the electronic frontier fund. Because that way you're working as a part of a polity and not just trying to, like, steer the market with consumption choices. You know, people who tell you to vote with your wallet typically have thicker wallets than you and anticipate winning that vote. Make consumption choices that make your life better if you want to do that, right? So, like you, I have a framework laptop. Like you, I'm a Linux user. I have all kinds of things in my life that make my life better. I use RSS. I go to my local bookstore and support them. I have a great mom and pop grocery store. If you're ever in Burbank and you want to go to a great little local grocery store, go to Handy Market. Amazing bourbon selection. They smoke tri-tips every Saturday. They've got a great deli counter. They've got incredible produce. Great selection of cheeses. A giant ice cream freezer. They have like three different great roasters in their coffee section. Go to your local grocery store if it makes your life better. But the minute you find yourself agonizing over this shit, the minute you're like, well, Twitter is the only way I can stay in touch with the people in the country I moved away from. Should I leave? Is it immoral for me to be on Twitter? Just cut it out. Do the thing that makes you happy. Stay on Twitter and spend the energy that you were going to spend agonizing over Twitter over Blue Sky and use it to work with your local hack space or work with your local solidarity group or work with your local mutual aid group or work with a neighborhood association or work with an EFF group.

And try and make a systemic change as part of a polity. Use it to organize a union. Use it to support a union. Use it to do things that, like, make a big systemic change. You know, there's a great section in Zephyr Teachout's book, Break Them Up, which is her book about monopolies, towards the end, where she says, if there is a demonstration against Amazon at the local Amazon warehouse, and you don't want to buy your markers from Amazon to make your sign because you think that would be like bad. And so you drive around for hours looking for an artisanal marker shop and you miss the protest. Jeff Bezos wins that round. Right. Fair. Yeah. You know, like, like, just, just keep your eye on the ball and just, and look, and, and like, I'm not saying that a boycott doesn't work. Boycotts can absolutely work, but a boycott isn't like you agonizing over your consumption choices. It's a polity. It is a group of people deciding en masse not to buy something, not to patronize something. You know, Martin Luther King and the bus boycott was not about people like, you know, checking their vibes and deciding which things they were and weren't going to do. It was about solidarity within a group making a collective decision. So yeah, boycotts work, but only when they're organized boycotts.

Great advice. Thank you so much, Corey. I appreciate it. I will look forward to that next two, three, four, five books coming out over the next year, whatever their release schedule looks like. And safe travels on the book tour. Thanks again, Corey. Thank you very much. Take care. Thank you.