Category Archives: Free speech

Elon Musk needs to learn that more debate does not mean more truth

Posted on by

Today’s Observer column:

Underpinning Musk’s views about free speech and the public sphere (AKA town square) is the fatuous metaphor of “the marketplace of ideas” that emerged from the deliberations of the US supreme court in 1953 (though something like it was mooted by Justice Oliver Wendell Holmes way back in 1919). It suggests that ideas compete with each other in a conceptual marketplace where they can be critically evaluated by every individual. As law professor David Pozen and others have pointed out, there’s no empirical evidence that a larger volume of speech, or a more open “marketplace” of ideas, tends to lead people away from falsity and towards truth. Subscribing to the metaphor is thus either a matter of faith or of evidence-free credulity. And if Musk believes that it is the secret sauce for managing Twitter then he’s a bigger crackpot than even I thought.

Do read the whole thing.

Sunday 12 July, 2020

Posted on by

A house full of music

This astonishing programme was broadcast on BBC1 this evening. It’s on the iPlayer for a month and is unmissable IMHO. Take an hour off and watch it.

Could juries be a solution to the free-speech moderation problem on social media?

This morning’s Observer column on Jonathan Zittrain’s big idea.

One of the most instructive experiences of my life was serving as a juror in a criminal trial. When the summons to report for jury service arrived, though, I was anything but enthusiastic. I was bringing up two young children on my own at the time and the last thing I needed was to be locked down for an unknown number of days. So I headed into the crown court feeling pretty glum.

The trial was a serious one: the charge was of causing grievous bodily harm with intent. It went on for two weeks. A number of witnesses gave evidence, much of which seemed (to me) unconvincing, sometimes contradictory, occasionally horrifying. We learned more about what goes on at night in an economically depressed East Anglian town than is good for anyone. And then, when the lawyers and the judge had summed up, we retired to reach a verdict.

What happened next was remarkable…

Read on

History of (tech) ideas?

One of the biggest treats of the lockdown has been David Runciman’s ‘History of Ideas’ podcast — a set of absorbing, enlightening and thought-provoking lectures on some of the intellects who have shaped the way we think about politics. In this first set of talks he covered:

  • Thomas Hobbes on power
  • Mary Wollstonecraft On Sexual Politics
  • Benjamin Constant on Liberty
  • Alexis de Tocqueville on Democracy
  • Marx and Engels on Revolution
  • Mahatma Gandhi on Self-Rule
  • Max Weber on Leadership
  • Friedrich Hayek on the market
  • Hannah Arendt on Action
  • Frantz Fanon on Empire
  • Catharine MacKinnon on Patriarchy
  • Francis Fukuyama on History

(The only one I would have added is John Maynard Keynes.)

If there had been a season of talks like this on the radio when I was a teenager I might have decided to study politics rather than engineering. I can imagine these podcasts having a similar impact on serious teenagers wondering what A-Levels to study now. What made the talks so good was the way they provided the context needed if one is embarking on reading, say, Hobbes or Weber for the first time.

Having really enjoyed the series, I then fell to wondering who would be the thinkers for an analogous series on computing and computation.

Here’s a first stab at such a list:

  • George Boole on logic
  • Charles Babbage and Ada Lovelace on automated calculation
  • Alan Turing on computability
  • John von Neumann on computer architecture
  • Norbert Weiner on automation
  • Claude Shannon on information theory
  • Donald Knuth on algorithms
  • Vannevar Bush on associative linking
  • JCR Licklider on computers as communication devices
  • Douglas Engelbart on augmentation
  • Paul Baran and Donald Davies on packet-switching
  • Ted Nelson and Tim Bernard-Lee on hypertext
  • Hal Varian on the economics of information goods
  • Stuart Russell on AI
  • Shoshana Zuboff on surveillance capitalism

Who am I missing? Nick Bostrom on superintelligence?

With India’s TikTok Ban, the World’s Digital Walls Grow Higher

As this NYT story illustrates, censorship and politics are fracturing the global internet, isolating users and industries accustomed to ignoring national borders.

TikTok, the first Chinese internet service to have a truly global fan base, is rapidly falling victim to China’s worsening diplomatic relations around the globe. It is yet another sign that the digital world, once thought of as a unifying space that transcended old divisions, is being carved up along the same national lines that split the physical one.

Tensions between India and China have run hot ever since a border clash in the Himalayas two weeks ago left 20 Indian soldiers dead. The government in New Delhi announced a ban on 59 Chinese apps late Monday, saying they were secretly transmitting users’ data to servers outside India.

India’s decision strikes at a number of China’s leading technology companies, including Alibaba, Tencent and Baidu. But perhaps none will be more affected than TikTok and its Beijing-based parent, ByteDance, which has built a huge audience in India as part of an aggressive and well-funded expansion around the world. TikTok has been installed more than 610 million times in India, according to estimates by the data firm Sensor Tower. In the United States, the app has been installed 165 million times.

‘Balkanisation’ of the Internet picks up speed.

Imagining New York without cars

Farhad Manjoo’s imaginative essay is worth some of your time. It’s a nice example of how to tell a complex story using Web technology imaginatively.

That whirring sound you will hear is that of Robert Moses rotating at 5,000rpm in his grave.

This blog is also available as a daily email. If you think this might suit you better, why not subscribe? One email a day, delivered to your inbox at 7am UK time. It’s free, and there’s a one-click unsubscribe if you decide that your inbox is full enough already!

Monday 6 July, 2020

Posted on by

A royal enclosure

On our cycle-ride this afternoon we passed Childerley Hall, a substantial country house dating from the early 17th century. It was the house where King Charles I was held overnight in June 1647 while being taken from Northamptonshire to Newmarket. Charles was executed two years later.

Click on the image for a larger version.

Tough times ahead for UK universities

From today’s Financial Times:

About a dozen universities could be at risk of insolvency because of the coronavirus pandemic, with a potential drop in the number of students and mounting pension liabilities leaving the sector facing an unprecedented financial crisis.

With losses across higher education totalling anywhere between £3bn and £19bn, or between 7.5 per cent and half of overall annual income, some institutions may be unable to survive without a targeted government bailout, according to research by the Institute for Fiscal Studies think-tank…

Martin Wolf and the relevance of Aristotle

It is clear then that the best partnership in a state is the one which operates through the middle people, and also that those states in which the middle element is large, and stronger if possible than the other two together, or at any rate stronger than either of them alone, have every chance of having a well-run constitution.”

  • Aristotle, Politics

Martin Wolf has a wonderful essay in today’s FT in which he applies Aristotle’s insight into the recent history of liberal democracies. I’m not sure if the link is behind a paywall today, but recently the FT has been good about lowering the wall on pieces like this which enrich the public sphere.

239 Experts With One Big Claim: The Coronavirus is airborne

Really important report.

The coronavirus is finding new victims worldwide, in bars and restaurants, offices, markets and casinos, giving rise to frightening clusters of infection that increasingly confirm what many scientists have been saying for months: The virus lingers in the air indoors, infecting those nearby.

If airborne transmission is a significant factor in the pandemic, especially in crowded spaces with poor ventilation, the consequences for containment will be significant. Masks may be needed indoors, even in socially-distant settings. Health care workers may need N95 masks that filter out even the smallest respiratory droplets as they care for coronavirus patients.

Ventilation systems in schools, nursing homes, residences and businesses may need to minimize recirculating air and add powerful new filters. Ultraviolet lights may be needed to kill viral particles floating in tiny droplets indoors.

The World Health Organization has long held that the coronavirus is spread primarily by large respiratory droplets that, once expelled by infected people in coughs and sneezes, fall quickly to the floor.

But in an open letter to the W.H.O., 239 scientists in 32 countries have outlined the evidence showing that smaller particles can infect people, and are calling for the agency to revise its recommendations. The researchers plan to publish their letter in a scientific journal next week.

It’s frustrating to see how slowly the world converges on obvious common sense. For the foreseeable future the only way of having a hope of some kind of normality is for people to wear masks whenever they’re in an indoor space with non-family members. How long will it take for this penny to drop.

Use juries on social media decisions

Jonath Zittrain has had a good idea:

No matter how Facebook and its counterparts tweak their policies, whatever these companies do will prompt broad anxiety and disapprobation among experts and their own users. That’s because there are two fundamental problems underlying the debate. First, we the public don’t agree on what we want. And second, we don’t trust anyone to give it to us. At one moment someone can reasonably question how Facebook thinks itself king, deciding what each of its 2.4 billion users can see or post—and at another ask how Facebook can make a profit of $22 billion a year from distributing others’ content, and yet take no consistent responsibility for what’s inside it.

What we need are ways for decisions about content to be made, as they inevitably must be when platforms rank and recommend content for us to see; for those decisions yet not to be too far-reaching or stiflingly consistent, so there is play in the joints; and for the deep stakes of those decisions to be matched by the gravity and reflectiveness of the process to make them. Facebook recently announced plans for an “independent oversight board,” a tribunal that would render the company’s final judgment on whether a disputed posting should be taken down. But far more than its own version of the Supreme Court, Facebook needs a way to tap into the everyday common sense of regular people. Even Facebook does not trust Facebook to decide unilaterally which ads are false and misleading. So if the ads are to be weighed at all, someone else has to render judgment.

In the court system, legislators write laws, and lawyers argue cases, but juries of ordinary people are typically the finders of fact and judges of what counts as “reasonable” behavior. This is less because a group of people plucked from the phone book is the best way to ascertain truth—after all, we don’t use that kind of group for any other fact-finding. Rather, it’s because, when done honorably, with duties taken seriously, deliberation by juries lends legitimacy and credibility to the machinations of the legal system.

I think this is a great idea. I was once a juror on a really serious criminal case and I was blown away by the way in which a group of twelve ‘ordinary’ people was able to sort through a maze of contradictory and flawed evidence and patiently and thoroughly come to a decision. And then to discover afterwards that it was exactly the correct conclusion, because the previous criminal record of the accused had — of course — not been disclosed to us.

Of course juries sometimes get it wrong. But so do judges, sometimes. No system is perfect, but the jury system has been going for centuries and it has served law-abiding societies well.

How to run a Zoom webinar

If you have any aspiration or need to run a Zoom webinar competently and successfully (and I’ve seen a few car-crash examples of how not to do it in recent weeks) then this comprehensive check-list by my friend Quentin is a must-read. Pass the link on to anyone you think might need it.

All Harvard University students will take online classes this fall amid coronavirus

Some students will live on campus but all lectures will be online.

So reports the Boston Herald.

Harvard University is allowing some students to live on campus this fall amid coronavirus, but all classes will be taught online, the university announced on Monday.

“… All course instruction (undergraduate and graduate) for the 2020-21 academic year will be delivered online,” Harvard officials wrote to the campus community. “Students will learn remotely, whether or not they live on campus.”

While most students will live at home, Harvard plans to bring up to 40% of its undergraduates to campus, including all first-year students.

“This will enable first-year students to benefit from a supported transition to college-level academic work and to begin to build their Harvard relationships with faculty and peers,” the officials wrote. “Both online and dorm-based programs will be in place to meet these needs. Over the last few weeks, there has been frequent communication with our first-year students about their transition to Harvard and this will continue as we approach the start of the academic year.”

Oddly enough, there’s no mention of a reduction in tuition fees. So as one American university student commented to the WSJ the other day (and it’s not clear what university he had in mind): “It’s like paying $75000 for a ticket to see Beyoncé and then being told you’ve got a livestream instead”. Neat way of putting it, eh?

I hate to say it, but all this was foreseen by Eli Noam in his wonderful paper, “Electronics and the Dim Future of the University”. When was it published? Why, 1995.

How Trump can lose the election and still cling to power

My friends think I’m neurotic, but I can’t see Trump accepting defeat and going quietly. And when I ask “Well, who exactly would remove him — if need be physically — from the White House?” I don’t get an answer. So you can see why this piece in Newsweek grabbed my attention.

This spring, HBO aired The Plot Against America, based on the Philip Roth novel of how an authoritarian president could grab control of the United States government using emergency powers that no one could foresee. Recent press reports have revealed the compilation by the Brennan Center at New York University of an extensive list of presidential emergency powers that might be inappropriately invoked in a national security crisis. Attorney General William Barr, known for his extremist view of the expanse of presidential power, is widely believed to be developing a Justice Department opinion arguing that the president can exercise emergency powers in certain national security situations, while stating that the courts, being extremely reluctant to intervene in the sphere of a national security emergency, would allow the president to proceed unchecked.

Something like the following scenario is not just possible but increasingly probable because it is clear Trump will do anything to avoid the moniker he hates more than any other: “loser.”

Trump actually tweeted on June 22: “Rigged 2020 election: millions of mail-in ballots will be printed by foreign countries, and others. It will be the scandal of our times!” With this, Trump has begun to lay the groundwork for the step-by-step process by which he holds on to the presidency after he has clearly lost the election…

The authors (a former US Senator for Colorado and a senior journalist who was once senior counsel to a congressional committee) then lay out a 12-stage scenario for how the nightmare could unfold.

Worth reading unless you are of a nervous disposition.

This blog is also available as a daily email. If you think this might suit you better, why not subscribe? One email a day, delivered to your inbox at 7am UK time. It’s free, and there’s a one-click unsubscribe if you decide that your inbox is full enough already!

Saturday 27 June, 2020

Posted on by

Quote of the Day

Other countries are used to loathing America, admiring America, and fearing America (sometimes all at once). But pitying America? That one is new.

The dead tree

On one of our cycle routes. Dead trees make for very dramatic photographs, sometimes. I’m always tempted to stop and photograph them.

Can this really be right?

From today’s Guardian:

The UK government’s plan to invest hundreds of millions of pounds in a satellite broadband company has been described as “nonsensical” by experts, who say the company doesn’t even make the right type of satellite the country needs after Brexit.

The investment in OneWeb, first reported on Thursday night, is intended to mitigate against the UK losing access to the EU’s Galileo satellite navigation system.

But OneWeb – in which the UK will own a 20% stake following the investment – currently operates a completely different type of satellite network from that typically used to run such navigation systems.

“The fundamental starting point is, yes, we’ve bought the wrong satellites,” said Dr Bleddyn Bowen, a space policy expert at the University of Leicester. “OneWeb is working on basically the same idea as Elon Musk’s Starlink: a mega-constellation of satellites in low Earth orbit, which are used to connect people on the ground to the internet.

“What’s happened is that the very talented lobbyists at OneWeb have convinced the government that we can completely redesign some of the satellites to piggyback a navigation payload on it. It’s bolting an unproven technology on to a mega-constellation that’s designed to do something else. It’s a tech and business gamble.”

If true, it looks like Trump-level imbecility.

Simon Kuper on why football matters

Lovely essay:

I’m British but I grew up mostly abroad, so when I went to university in England I discovered a new species of man: the Total Fan, the teenager whose main identity was the football club he supported. I witnessed conversations in the common room that went like this:

Student in plastic Manchester United shirt: “We’re brilliant this season.”

Student in Spurs shirt: “No, you’re shit.”

Student in Crystal Palace shirt: “He’s right, Steve. You’re shit.”

They weren’t exactly casting aspersions on Steve’s personality. They were talking about his football club. However, they saw the two things as essentially the same. Steve was Manchester United. The Spurs fan once told me that, when his team won the FA Cup, he walked into the common room to receive everybody’s congratulations as if he personally had lifted the trophy.

Even if you’re not a football fan (and I’m not) this is worth reading.

Share the wealth as we recover health (hopefully)

Noema magazine (a new publication from the Berggruen Institute) has an interesting conversation with Joe Stieglitz and Ray Dalio about how to ensure that the benefits of any recovery from the Covid crisis are shared with the population as a whole.

The basic idea: the massive taxpayer-financed cash infusion to save some of the largest companies that are otherwise viable may present a unique opportunity to more effectively tackle inequality by bolstering the assets of the less well-off. If the same taxpayers who are bearing the costs of the bailout also share an upside when we recover prosperity, wealth will be shared more fairly.

“This can be done”, says the magazine,

by establishing a sovereign wealth fund, or national endowment, that pools the taxpayer’s ownership shares from all the bailed-out companies and distributes regular dividends to all citizens. We call this “universal basic capital,” as distinct from the idea of a universal basic income. Instead of only once again relying on redistributing income to close the gap after wealth has been created, that wealth should be shared upfront in what we call “pre-distribution.”

There are many models out there that guide us on this path. Alaska has long had a social wealth fund that pays dividends to citizens from the revenues of the state’s oil leases. Norway has a similar fund, also from oil revenues, that pays into the general pension system. Australia has what is calls the superannuation fund, in essence a sovereign wealth fund financed by employees, employers and state contributions for its universal pension scheme. The wealth of that fund now stands at almost $2 trillion, a sum greater than Australia’s GDP. Singapore has a similar plan, called the Central Provident Fund, from which citizens can also draw for health and housing needs. It is so profitable from its global investments that it is even able to fund some government services and help keep taxes low.

What is important at this point is to recognize the opportunity for reducing social inequality that can be created by a fair and innovative approach to economic recovery. If everyone in this pandemic must share the downside, all must share in the upside as well.

Some promising ideas here. And the good thing is that none of the corporations in which governments might take a stake in return for support during the pandemic are tech companies, for the simple reason that those companies are the ones that will have benefited most from the crisis.

Noema‘s good, btw.

We can make you hurt if you don’t do what we want

Jonathan Zittrain is my idea of a perfect academic. Staggeringly bright, knows both digital tech and the law intimately (he has Chairs in both Harvard Law and Engineering), fizzes with original and often productive ways of viewing tricky problems, etc. So whenever he writes or lectures about anything I pay attention.

Now he has an article in The Atlantic about what social media outfits should do about Trump. At the beginning, his discussion of the possible options for regulating the speech of an authoritarian nutter takes a fairly standard detached, scholarly tone. His emerging conclusion seems to be that every plausible configuration of social media in 2020 is unpalatable.

But then, he briefly switches to a different register:

Those proposals can be analyzed and judged on their own terms as if they simply appeared on Congress’s docket out of nowhere, and I’d normally offer here some thoughts on their details. But I can’t stay in my academic lane. The executive order, and the push for more legislation, is part of a larger pattern in which the president appears to seek vengeance against those who even mildly criticize him, retaliating in any way he can, including by using the powers of his office. When, for instance, he didn’t like The Washington Post’s reporting about him, he made it clear—on Twitter, fittingly enough—that, because the paper is owned by Amazon CEO Jeff Bezos, he would like to disadvantage Amazon however he can, including by demanding that the U.S. Postal Service raise its shipping rates. Here, the executive order is so scattershot, and the legislation so crudely sweeping, that it’s important to recognize that it conveys more than its text says. What it really says is: We can make you hurt unless you do what we want, and what we want is what helps the president personally. [Emphasis added.]

Yep: full marks. That’s the nub of Trump’s authoritarian threat. Same as Erdogan, Orban, Bolsonaro & Co.

So what does Zittrain think we should do?

“In the near term, the simplest solution is to vote Trump out of office.”

Well, yes: but you don’t have to be a bi-Chaired Harvard prof to come to that conclusion.

What if that option doesn’t work?

“In the longer term”, says Zittrain,

the most promising path for online content moderation lies in taking up unavoidable decisions by the largest companies in ways that respect the gravity of those decisions — likely involving outside parties in structured, visible roles — and, even more important, in decentralizing the flows of information online so that no one company can readily change the map.

So when Twitter tempers its deference and wades into a fraught zone by fact-checking in its own voice, still judged in the public sphere by its attention to the real facts, I respect its decision. One way to try to break what is raging behavior even—and especially—by a president is to create policies to deal with it, policies that would collect dust if the rule of law and the institutions designed to reinforce it were not under such extraordinary and explicit attack.

Yeah, sure. But this seems a bit feeble after the build-up. What might those policies look like? And how might we ‘decentralize’ those information flows?

Maybe there’s a sequel to this piece coming. If so I can’t wait.

Quarantine diary — Day 98

Link

This blog is also available as a daily email. If you think this might suit you better, why not subscribe? It’s free. One email a day, delivered to your inbox at 7am UK time. And there’s a one-click unsubscribe button if you decide that your inbox is crowded enough already!

Silicon Valley ideology in a nutshell

Posted on by

The proprietor of N-gate is an engineer who grew up in Palo Alto and now lives in the Pacific Northwest, where he works in high-performance computing. He agreed to exchange e-mails on condition of anonymity. “Almost every post deals with the same topics: these are people who spend their lives trying to identify all the ways they can extract money from others without quite going to jail,” he wrote. “They’re people who are convinced that they are too special for rules, and too smart for education. They don’t regard themselves as inhabiting the world the way other people do; they’re secret royalty, detached from society’s expectations and unfailingly outraged when faced with normal consequences for bad decisions. Society, and especially economics, is a logic puzzle where you just have to find the right set of loopholes to win the game. Rules are made to be slipped past, never stopping to consider why someone might have made those rules to start with. Silicon Valley has an ethics problem, and ‘Hacker’ ‘News’ is where it’s easiest to see.”

From a terrific New Yorker essay on the task of moderating Hacker News.

What Trump doesn’t know

Posted on by

From Kara Swisher:

I’m sorry to be the one to have to tell the president, but someone has to: Social media is not the public square, not even a virtual one.

Not Facebook. Not Reddit. Not YouTube. And definitely not Twitter, where a few days after Facebook announced it was barring some extremist voices like Alex Jones, President Trump furiously tapped out: “I am continuing to monitor the censorship of AMERICAN CITIZENS on social media platforms. This is the United States of America — and we have what’s known as FREEDOM OF SPEECH! We are monitoring and watching, closely!!”

He can monitor (yes, that’s definitely a creepy word) and watch all he wants, but it will not matter one bit. Because the First Amendment requires only that the government not make laws that restrict freedom of speech for its citizens…

Getting things into perspective

Posted on by

From Zeynep Tufecki:

We don’t have to be resigned to the status quo. Facebook is only 13 years old, Twitter 11, and even Google is but 19. At this moment in the evolution of the auto industry, there were still no seat belts, airbags, emission controls, or mandatory crumple zones. The rules and incentive structures underlying how attention and surveillance work on the internet need to change. But in fairness to Facebook and Google and Twitter, while there’s a lot they could do better, the public outcry demanding that they fix all these problems is fundamentally mistaken. There are few solutions to the problems of digital discourse that don’t involve huge trade-offs—and those are not choices for Mark Zuckerberg alone to make. These are deeply political decisions. In the 20th century, the US passed laws that outlawed lead in paint and gasoline, that defined how much privacy a landlord needs to give his tenants, and that determined how much a phone company can surveil its customers. We can decide how we want to handle digital surveillance, attention-channeling, harassment, data collection, and algorithmic decision­making. We just need to start the discussion. Now.

Quote of the Day

Posted on by

“The problem will never be solved, if solving it means getting rid of all the bad stuff, because we can’t agree on what the bad stuff is. Knowing that things won’t be perfect, what do we feel is most desirable? A system that errs on the side of caution, or one that errs on the side of being permissive?”

Rasmus Nielsen, Reuters Institute, Oxford.

WhatsApp tries damage limitation

Posted on by

This morning’s Observer column:

In the last two years, around two dozen people in India have been killed by lynch mobs inflamed by rumours on WhatsApp, the encrypted messaging service owned by Facebook. WhatsApp has also been fingered for its role in other hateful or unsavoury episodes in Brazil and Pakistan. In each case, the accusation is essentially the same: disinformation and lies, often of an inflammatory kind, are effortlessly disseminated by WhatsApp and obviously believed by some of the recipients, who are thereby encouraged to do terrible things.

In terms of software architecture and interface design, WhatsApp is a lovely system, which is why it is a favourite of families, not to mention Westminster plotters, who are allegedly addicted to it. Its USP is that messages on the platform are encrypted end to end, which means that not even Facebook, the app’s owner, can read them. This is either a feature or a bug, depending on your point of view. If you’re a user, then it’s a feature because it guarantees that your deathless prose is impenetrable to snoopers; if you’re a spook or a cop, then it’s definitely a bug, because you can’t read the damned messages.

A few years ago, WhatsApp added a key new feature – an easy way to forward a message to multiple chat groups at once…

Read on

Tyler Cowen on the impossibility of regulating speech on Internet platforms

Posted on by

From his latest Bloomberg column:

I’d like to suggest a simple trilemma. When it comes to private platforms and speech regulation, you can choose two of three: scalability, effectiveness and consistency. You cannot have all three. Furthermore, this trilemma suggests that we — whether as users, citizens or indeed managers of the platforms themselves — won’t ever be happy with how speech is regulated on the internet.

One view, which may appear cynical, is that the platforms are worth having, so they should appease us by at least trying to regulate effectively, even though both of us know they won’t really succeed. Circa 2019, I don’t see a better solution. Another view is that we’d be better off with how things were a few years ago, when platform regulation of speech was not such a big issue. After all, we Americans don’t flip out when we learn that Amazon sells copies of “Mein Kampf.”

The problem is that once you learn about what you can’t have — speech regulation that is scalable, consistent and hostile to bad agents — it is hard to get used to that fact. Going forward, we’re likely to see platform companies trying harder and harder, and their critics getting louder and louder.

I like his ‘trilemma’ idea. It reminds me of Dani Rodrik’s one, which says that democracy, national sovereignty and global economic integration are mutually incompatible: we can combine any two of the three, but never have all three simultaneously and in full.