This Will All Be So Great If We Don’t Screw It Up
Cory Doctorow on Big Tech and Pathways Forward
As the power, influence and scale of the major firms that dominate Silicon Valley continues to grow, the collective response might be best described as befuddled. From our mental health to influence on elections to copyright and fair business practices, the impacts are wide reaching and complex. Joshua Fouts sat down with Cory Doctorow to discuss the overall state of Big Tech at the moment. Doctorow is a prolific journalist, blogger, creative commons advocate, Electronic Frontier Foundation Fellow, and award-winning science fiction author whose latest book, RADICALIZED, was published in March.
Doctorow last spoke at a Bioneers Conference in 2017 – watch his keynote talk on The Fight for a Free, Fair and Open Internet here.
JOSHUA FOUTS: Cory Doctorow, thank you so much for joining me. This is a real pleasure and a thrill.
In some ways, these Big Tech companies can be thought of as the robber barons of the 21st century. Their reach is enormous and what to do about them is becoming a political talking point. In your May 2019 column in The Locus, “Steering with the Windshield Wipers,” you focused on the history of monopolies and monopoly regulation in the US at it relates to technology companies today. Given the recent political posturing around this topic from Elizabeth Warren and others, I’d love to hear your thoughts on the issue. How big is too big? What is to be done about the increasing size and scope of today’s major technology problems?
CORY DOCTOROW: I’ve given some thought to what I would like written on my tombstone. And what I would like written on my tombstone is something like: “This will all be so great if we don’t screw it up.” The kind of technology that I want to see in our future is one that is resilient on a technological and on a policy level because it is distributed and pluralistic. Right? Rather than having five giant websites filled with screenshots from the other four, I would like there to be a lot of different places where people accomplish a wide variety of goals online – communication, fellowship, commerce, any of the other things we do, romance and so on.
What I would like written on my tombstone is something like: “This will all be so great if we don’t screw it up. — Cory Doctorow on today’s technology problems.Tweet
The history of the Internet for most of its life is of a system that was very, very dynamic, where last year’s winners were overturned by new entrants so routinely that you could barely keep track of who it was. I remember a conference around 2002 or 2003 put on by Kevin Werbach featuring Sergey Brin from Google. He said that people didn’t need to worry about how big Google was getting because all it takes for someone to change search engines is to update their bookmarks, and when that happens, Google will go away. So the only way Google was going to remain on top was if their search engine was better than any other search engine a user might try.
Over time we’ve had this massive concentration in tech, and that has happened at the same time as we’ve had massive concentration in virtually every other industry you care to name: automotive production, energy, shipping and logistics, pro-wrestling, etc… John Oliver did this infamous, wonderful episode about how there used to be 30 pro-wrestling leagues and now there is one, and it has all the problems that you would expect to have with monopoly. It squeezes its suppliers, in this case the wrestlers, who are now non-consensually reclassified as contractors, which means that they’re no longer entitled to medical benefits, which means that these people who do this very physically dangerous thing are using Go Fund Me to pay for their healthcare, and they’re dropping dead in their 50s. Every industry has been attended by mass concentrations – we have four movie studios, five publishers, four record labels. Finance has gone through this. Pharma has gone through this.
When people talk about Big Tech, they repeat the same tech exceptionalist lines that tech itself uses to describe why it got so concentrated. They say Big Tech avails itself of network effects and of first-mover advantage, and sometimes of a semi-mystical capacity to use machine learning to make people addicted to a product. For these reasons, they say, tech is a winner-take-all phenomenon.
It’s a really interesting evolution in how we talk about tech because it wasn’t so long ago that we were talking about the long tail and we were saying that what tech enabled was this massively pluralistic marketplace where every conceivable idea could find its audience no matter how niche it was. Think about the initial shape of the blogosphere. I remember Technorati releasing an early report saying more than half of all blogs had three or fewer in-bound links from other blogs, which meant that more than half of all blogs were so niche that they were viable with an audience of three. Today, that is not the case.
Today we have this massive concentration and Big Tech – and a lot of critics – want you to think that it is because of these network effects. The thing that I find very unconvincing about this narrative is that pro-wrestling doesn’t have network effects, but it has undergone the same concentration. As an example, take eyewear: one company – Luxottica, the Italian company – owns every eyewear brand you’ve ever heard of. They also own every eyewear retailer you’ve ever patronized. They own LensCrafters and Sears Optical, and Sunglass Hut. Anytime an eyewear brand won’t sell their company, Luxottica will boycott the company in their retailers until the company is driven to its knees, and Luxottica can buy it for pennies on the dollar. They own every major lab that makes lenses and contact lenses, they own every major insurer that insures optical, and they own every eyewear brand and every eyewear retailer. They don’t have network effects. They don’t have first-mover advantage. They don’t have globalism. They just have oligarchy.
It behooves us, I think, to ask: Is there an alternate set of explanations that explains why Big Tech is giant, that also explains why big optical is giant? I think there is, and it’s that until the mid-‘80s, companies were not allowed to merge with their major competitors, called merger to monopoly. They also weren’t allowed to buy their small nascent competitors, the companies that might grow to challenge them. They had functional separation obligation. If you were a bank, you couldn’t own a company that competed with the companies you were lending money to because it would give you a huge advantage over them. If you were a rail company, you couldn’t own the freight companies that put the freight in your rail cars because you’re competing against the other freighters who wanted to put freight in your rail cars.
Ronald Reagan was a great fan of Robert Bork, a hobbyist economist and lawyer (I call him an “alternate historian”). Bork believed that if you squinted really hard at the legislative history of antitrust laws, the Sherman Act in the US, that the congressmen who debated this (and it was congressmen in those days) were totally indifferent to the idea of monopoly, and that the only thing that the Sherman Act really prohibits is what’s called consumer harm. That is when you merge two companies and in the short term they raise prices on consumers. Not in the long term, because Luxottica has raised the price of eyewear a thousand percent relative to where it was before they started, but in the short term. If you could show that there wouldn’t be immediate price increases, then you could have any kind of merger, any kind of anti-competitive activity, any kind of acquisition.
Tech companies grew by doing the things that Robert Bork fought to legalize, and that Reagan ultimately legalized (and that his successors – Bush I, Bush II, Clinton, and Obama – all enabled.) They bought their competitors, they merged with their biggest competitors, and they vertically or horizontally integrated so that they would have the rail cars and the freight, so that they could be the lender and the business that took the loan.
Google has only ever had two successful products that they developed in-house – Gmail, which is a clone of Hotmail, and Search. Everything else Google does came from an outside acquisition. When you think through possible alternate explanations, you have to ask yourself: Is it possible that the reason we have monopolies is that we took away the rules that we used to have to prevent monopolies? You can look at those monopolies and realize that everything they did to become a monopoly is a thing that used to be banned under anti-monopoly rules. Maybe the rules were working. Maybe it wasn’t the mystical power of machine learning to turn us into addicts. Maybe it wasn’t network effects. Maybe it was just that we had rules that worked and then we stopped employing them.
Ask yourself: Is it possible that the reason we have monopolies is that we took away the rules that we used to have to prevent monopolies? – Cory DoctorowTweet
Listening to true believers in Borkian-Reaganite market capitalism explain why it must be machine learning or globalism or network effects is like listening to a lifelong smoker who’s finally got lung cancer who wants you to understand that the reason they have lung cancer is not because they unwisely kept jamming cigarettes in their face, it’s because of environmental toxins.
So this brings me to Liz Warren. Full disclosure: I am a great supporter of Elizabeth Warren and am a donor, both to this campaign and her previous campaigns. I’m Canadian (we’re like serial killers, we’re everywhere, we look like everyone) and I’m eligible to become a citizen during the next presidential administration. If I could walk into a federal office and put my hand over my heart and swear allegiance to the flag under the grinning face of President Elizabeth Warren, I would be a very, very, very happy man.
That said, I think that we need to understand that we can try to fix Big Tech or we can try to fix the Internet, but we can’t do both. So fixing Big Tech would be like saying if you’re a rail company and you want to carry freight, this is the price schedule that you have to set forth. Everyone has to get those prices. Or you have to make sure that you don’t discriminate against these kinds of freight companies. Or you might say if you’re a banker and you want to lend money to businesses that you own, this is the way that you have to manage that separation, this is the wall you have to erect between your two business units. Those compliance costs are very expensive, and they’re very complicated, and they have a pretty low chance of working because in order to make them work, you have to be privy to things happening inside board rooms that are not necessarily visible even to an in-depth audit. They require constant and near-total scrutiny.
A lot of what has been proposed for Big Tech amounts to a very complex regulatory regime to make them behave themselves as “good monopolies.” Take Facebook or Twitter and the issue of online harassment and violent or extremist content. We say to them, you have to have some kind of regime where you can detect and eliminate these activities on your platform. You need to build filters, you need to hire rooms full of people. Or we say that since the livestream of the mosque shootings in Christchurch on Facebook meant that 2.3 billion people can see it, that Facebook now have to have some mechanism to determine whether every livestream on Facebook is or isn’t a mosque shooting.
We say that because YouTube captures all of the videos in the world now means that they need to spend at least $100 million to filter everything that people upload to YouTube under a new European copyright rules. What these regulations do is they lock in the permanent advantage of the Big Tech companies because they say that you cannot launch a rival to YouTube unless you have $100 million to build a YouTube filter.
There are distinctions made where if you’re a small Facebook competitor, you have a different duty than if you’re a large Facebook competitor, but getting that right is really hard. What it tends to do is produce glass ceilings where, for example, during the copyright directive fight in Europe, at one point there was a rule that said you didn’t need filters until you turned over 10 million euros in one year. That means that the day you earn your 10 million and first euro is the day you have to have 100 million euros to build the filter to comply with the regime.
You can try to come up with a gradient, but ultimately all of these things cut against the other possible remedy we have for Big Tech, which is breaking them up, which is what Warren also suggests. I’m very sympathetic to breaking up Big Tech. I’d like to see them broken up on functional lines, I’d like to see ad technology firms left out of the social media and search business, I’d like to see them broken up on merger-to-monopoly lines, I want to see Facebook divesting itself from Instagram.
If we say to be Facebook you need to be big enough to buy all these filters, then we also say we will never make Facebook so small that it can’t afford these filters. You can’t do both. We know how this works because the last company that ever faced meaningful antitrust remedies in this country was AT&T. AT&T had resisted breakup for generations because every time someone said let’s break them up, AT&T said, “Let us introduce you to the stack of public safety duties that Congress expects us to fulfill, that the FCC expects us to fulfill. Fulfilling these requires that we be as big as we are.” As soon as you give state-like duties to a firm, you effectively guarantee that you will never make that firm too small to serve as a de-facto arm of the state.
I’m in favor of vigorous regulatory enforcement. I want the FCC and the FTC to regulate the Big Tech and telecom companies. I don’t want those companies to self-regulate because they suck at it, because they are permanently compromised, no matter how much they say, “We have this arm’s-length person in the firm who’s paid by us, who doesn’t answer to us, and they will act as the regulators proxy.” We know how that works because every one of the big four accounting firms has been embroiled in a series of horrific scandals, each worse than the last in the last two or three years, in which that whole mechanism has totally broken down because they get paid by the company whom they are responsible for keeping honest, and ultimately you cannot be paid by the company you’re responsible for keeping honest because your divided loyalty will screw you.
When people hear, “Don’t regulate Big Tech because it will give them a permanent advantage,” they think that I’m saying, “Don’t regulate Big Tech at all.” What I’m saying is regulate Big Tech first by making them small enough that we can hold them to account, and second by making them small enough that we can make them pay their taxes so that we have the money to pay the regulators to hold them accountable. Anything less than that is just enshrining their permanent dominance. It’s effectively replacing the dream of a democratic tech with a constitutional monarchy where we say to these firms, King Zuck, King Sergey, King Larry, you guys get to rule the Internet forever, but there is an aristocratic parliament who will meet, a house of lords, and they will drape you with golden chains that you will suffer, and those golden chains will bind your behavior in ways that will reflect what the aristocracy thinks you should do, but you will never be answerable to the unwashed masses who will only have one social network, one search engine, where they go to conduct the democratic business of using the nervous system of the 21st century. I think that that is throwing in the towel.
FOUTS: Against the backdrop of all of this and the focus on the downside of Big Tech, are there things you’re excited about or that we should we excited about in the larger field at the moment?
DOCTOROW: That’s the “this will be so great” part (if we don’t screw it up.)
I can’t imagine a future where we survive the climate emergency, or find justice for people who’ve been historically discriminated against and suffered the worst of environmental problems and other forms of economic and coercive violence, that doesn’t involve using Big Tech or at least tech, but it requires that it be a democratic tech.
If you are looking at Puerto Rico post-Hurricane Maria and saying we need to take power generation out of the hands of giant corporations that are looking for the fastest buck with the lowest cost, and return it to community control, we need to create sustainable, renewable micro-grids. These micro-grids would form federated agreements with one another to share across one community to the next, where the profits remaining in the community can be used to do maintenance, upgrading and so on.
One of the reasons we have big centralized power generation is because it requires huge amounts of logistical oversight and coordination that you can only accomplish with computers, with really advanced embedded systems, and also systems that are in the hands of both the users and the technicians that can monitor the health of the system and tune it in real time.
I just read a great paper about how connected fields of wind-based power turbines can tune their wake, their backwash, so that as they turn the blades, they tune the whole turbine so that the turbulence that they create maximizes the speed of the turbine behind it through intense, real-time computation, looking at things like the fluid dynamics of wind in real time.
It’s not just that we’re able to do these things, it’s that we’re really going to have to. We are going to have to be able to do things like use computation to measure the outcomes of policing to determine whether or not our police are practicing racial discrimination, whether overt or intrinsic racial bias, and the only way we’re going to do that is with data. Take predictive policing which uses policing data and directs future police activity using a predictive algorithm.
If you gather data from racist policing practices and use it as your ground truth, the crime is only where you look for it. The only place where you find weed in people’s pockets is the neighborhood where you stop everyone and make them empty their pockets. By definition you don’t detect weed in people’s pockets in neighborhoods where no one has to empty their pockets. So if you only ever make brown people in brown neighborhoods empty their pockets, the algorithm will predict that that’s where all the weed is going to be. What this does as presently construed is reify and produce a veneer of objectivity to existing racist practices.
The Human Rights Data Analysis Group did work analyzing police data and comparing it to other data and were able to validate or invalidate the predictions. Turns out you can use this same data set and algorithmic approach, with a few tweaks, to actually identify whether or not police are practicing discriminatory or non-discriminatory policing. It is literally the same technology, it just vests the locus of control in different hands.
I think the problem with Big Tech is where the locus of control is. It’s that in order to talk to all of your friends online, you have to commodify your relationship through Facebook advertising algorithms. That the search algorithms run by Google use secretive tools to do ranking that make it hard to discover when the search result that you’ve been served is or isn’t trustworthy. It’s not subjected to peer review the way other information that we want to trust in our world is.
The “This will be so great” part is actually not so hard to distinguish from the “If we don’t screw it up” part. Because the difference is who has control over the system. Is it the people who use the systems? Or is it the people who are trying to extract value from people who use the system? That’s really the major difference.
FOUTS: My final question is: Is there anything you’ve read lately that you recommend others pick up in and around things that get you excited?
DOCTOROW: What’s good recently? I’ve nearly finished Naomi Klein’s next book, which is called On Fire, and it’s a collection of her essays from about the past decade or so. Watching the way that we think about the climate emergency changes over that decade is super instructive. In some ways it gives me hope because you can see the Overton window opening. Right? You can see the things that we’re allowed to say about climate without worrying that we’ll be laughed out of the room expanding.
I really liked Karl Schroeder’s new book Stealing Worlds. It is a very, very good book about climate emergency, civil liberties, and the subversion of totalitarian computer networks to achieve anti-totalitarian, pro-egalitarian, anti-oligarchic ends. It’s a cracking adventure novel and really fun to read.
There’s a really good, very dark book that just came out from Richard Kadrey called The Grand Dark, and it’s a kind of Weimar, dieselpunk novel. It’s set in this inter-war period where there’s this hectic energy, the war has ended, they can hear the drums of war in the distance, and everyone is in this kind of frenetic sex-drugs-and-rock’n’roll period in this fictional town, but the town is full of the mamed soldiers from the last war, and then also the automata, the robots that were developed in the last war, which are creating this kind of army of disaffected, unemployed people who are clearly being teed up to be cannon fodder. The robots are also developing genetically engineered chimeras that they called eugenics, that are being sold as pets but are the descendants of the war dogs that were bred for the previous war. Clearly the same labs that are turning out these pets are also turning out the next generation of biological, synthetic-biological organisms to fight the next war. It resonates very beautifully with our current moment. Kadrey is so good at writing hard boiled. It’s like someone took like a Tom Waits album, some really peaty Scotch, and a really shitty cigar and put them in a nutribullet and you’re drinking it and it’s terrific.
FOUTS: What’s your latest writing?
DOCTOROW: I have three books now in the pipeline. The one that’s through production and now in the schedule for 2020 is a picture book for little kids called Poesy the Monster Slayer about a little girl who is obsessed with fighting monsters, and she takes apart the super girly toys in her room and repurposes them into monster-slaying weapons. So her Barbie Dreamhouse roof becomes a shield and her Barbie bubble gum perfume becomes mace that she uses, her tiara is the silver she uses to repel the werewolves and so on.
I’ve got a novel for adults coming – not on the production schedule yet but it will be soon. It’s the third Little Brother novel.
I’ve just finished a very short (25,000-word) nonfiction book. It’s a long essay about competition and surveillance capitalism, why machine learning is a problem, how to understand the debate about surveillance capitalism in the framework of monopoly capitalism, and how those two are interrelated. It’s called Working as Intended: Surveillance Capitalism Is Not a Rogue Capitalism. It is with my publisher now.
FOUTS: Fantastic. Cory Doctorow, thank you so much for your time. Thanks for all the work that you do. It’s really, really incredible and vital against the backdrop of 2019 Planet Earth.
DOCTOROW: Well thank you, and thanks for the work Bioneers is doing and you’re doing. You guys are amazing.