Nieman Foundation at Harvard
Postcards and laundromat visits: The Texas Tribune audience team experiments with IRL distribution
ABOUT                    SUBSCRIBE
Jan. 22, 2020, 10 a.m.

Public infrastructure isn’t just bridges and water mains: Here’s an argument for extending the concept to digital spaces

“Our solutions cannot be limited to asking these platforms to do a better job of meeting their civic obligations — we need to consider what technologies we want and need for digital media to have a productive role in democratic societies.”

Our friend and colleague Ethan Zuckerman has written an important piece I’d like to draw your attention to. In a piece for the Knight First Amendment Institute at Columbia, he lays out what he calls the case for digital public infrastructure. (He also published a summary of it here; the main piece is kinda long.)

Most governments have traditionally argued that there are certain critical societal assets that should be built, managed, and controlled by public entities — think streets, airports, fire fighting, parks, policing, tunnels, an army. (And in just about every rich country except this one, access to and/or the provision of health care.) The choice to have, say, a city-owned park reflects two key facts: first, a civic judgment that having green outdoor spaces is important to the city; and second, that free parks open to all are unlikely to be produced by private companies driven by a motive for profit.

When it comes to the Internet we all live on, huge swaths of it are owned, controlled, and operated by private companies — companies like Facebook, Google, Amazon, Apple, Microsoft, and Twitter. In many cases, those companies’ public impacts aren’t in any significant conflict with their private motivations for profit. But in some cases, both Ethan and common sense argue, they are. Is there room for a public infrastructure that can offer an alternative to (or reduce the harm done by) those tech giants?

Here’s Zuckerman:

The introduction of new media technologies invariably brings about potent social and economic shifts. We are well into one of those shifts as the advent of the consumer internet has destabilized existing models of news production and distribution and enabled new hegemons to establish massive and powerful businesses. Two decades into this shift, societies are asking difficult questions about whether internet technologies and the business models that accompany them are dangerous for our citizens and our democracies.

At these moments of technological shift, it’s easy to assume that the business models adopted by technological innovators are inevitable and singular. They are not. As Paul Starr established in his magisterial The Creation of the Media, the paths taken by different nations in their adoption of new communication technologies (movable type, postal mail, telegraph, radio) depend on the politics and economics of the nation as a whole and vary widely from country to country.

This variance continues with the internet, even though the dominance of the United States — and Silicon Valley in particular — creates the illusion that a single economic and legal system governs our online spaces. This illusion obscures possible solutions to the challenges arising around the socially corrosive effects of new media technologies. Because we see the dominance of the internet by Google, Facebook, and others as inevitable, the solution space we consider for combatting mis-/disinformation, polarization, and promotion of extremism is overly constrained. Our solutions cannot be limited to asking these platforms to do a better job of meeting their civic obligations — we need to consider what technologies we want and need for digital media to have a productive role in democratic societies.

In other words: Being able to find things easily on the Internet is really important, to individuals, to businesses, to society at large. Google makes a really good search engine in a lot of ways. But it operates under its own set of incentives, generally around displaying ads that look a lot like search results but aren’t. Is search important enough to society for the government to say: “Instead of trying to convince Google to change its ways, we should provide an alternative that isn’t focused on selling ads”?

Or: A structured network that allows people online to connect with their friends and exchange news, pleasantries, or other information with one another is obviously important to everyone. Facebook has built a remarkable thing, a network of billions that has as good a graph of human relationships as any. People spend hours at a time using Facebook’s platforms and apps. But it also has serious negative effects on society and politics — whether that’s misinformation, radicalization, election interference, surveillance-as-advertising, or whatever your biggest complaint is. Is a social network important enough to society for the government to say: “Instead of trying to convince Mark Zuckerberg to change his ways, we should provide an alternative that isn’t coarsened by these awful outcomes that seem baked into Facebook’s very being”?

(I fully realize that about 85 percent of American readers are thinking THIS IS ALL INSANE right now. Stick with it.)

Among the ideas Zuckerman says are worthy of investigation:

  • Putting a tax on “highly surveillant advertising,” which he defines as “advertising that incorporates user tracking, combines demographic and psychographic data to create user profiles, or targets using factors other than a user’s stated intentions and geography.” That could produce, say, $2 billion a year that could be funneled to create and support public service digital media.
  • Using some of that money to build “auditable and transparent search and discovery tools.” Building a public version of Google is a bit ambitious, but he proposes starting by taking on a specific subfield or two of search. The goal is not to somehow overtake Google as the No. 1 search engine — it’s to provide an alternative for users and a testbed to learn things that could then be used to improve Google, too.
  • And while he’s at it, he also proposes a “non-surveillant advertising network…The popularity of podcast sponsorship, which offers far less information about user behavior, but associates ads with high-quality content, suggests that there might be a market opportunity for this model and that a non-surveillant ad network could ultimately be self-sustaining or capable of cross-subsidizing other projects.”

(He also calls for “support for the systemic study of the social and individual effects of digital media,” aimed at finding definitive answers to questions like “the addictive effects of social media, the idea that social media is increasing political polarization and isolating people in ideological echo chambers, that social media might push vulnerable individuals towards extremist political views and that targeted advertising might be so effective that it subverts democratic processes.” Which is a very good idea, to be sure — but c’mon Ethan, it’s a little gauche to make one of your recommendations “GIIIIIIIIIIVE MEEEEEEE MONNNNNNEEEEEEY.” 😀 )

This likely all sounds very…European to the Americans in the crowd. Public anything can be a hard sell in the United States, which throughout the rise of radio and television declined to launch a strong, BBC-style public broadcaster, instead letting private media companies divvy up the market among them.

(PBS and NPR didn’t hit the air until 1970 and 1971, respectively, long after TV and radio dials were first populated. The U.S. still spends only a tiny fraction of what other democracies do on public media. Canada spends 10× as much as the U.S. does per capita on public media; Japan spends 18×, Australia 19×, France 24×, the U.K. 38×, Germany 47×, and Norway 60× as much.)

But it’s common even in this country for sectors of life to begin through private, profit-driven experimentation and then — once the nascent sector becomes critical to American life — to become more public, either through regulation or public ownership. Many of the earliest roads in the United States were privately owned turnpikes, with tolls paid by the travelers using them. The same was true of many bridges and ferry crossings. Eventually, state and federal governments determined that the transportation network was too important to be left to private profit incentives and should be predominantly public-owned. The same was true of railroads, electricity, postal service, telephone service, and more — each initially delivered by the private sector, and each over time became either publicly regulated or publicly provided in some important way.

It is very easy to picture the line at your local DMV and imagine a worst-case scenario for “public digital infrastructure” — a Government Google that works as well as the first days of, the feds being a central node in your digital network of personal relationships, and so on. And anything public in this space would face a host of challenges, from privacy concerns to funding challenges to regulatory capture and more.

But that said — I agree with Ethan that this is a space increasingly worth thinking about (and doing something with). The enormous power over our lives held by those few giants in Silicon Valley is almost unprecedented in American corporate life; even Standard Oil, U.S. Steel, and Ma Bell mostly flexed their monopoly power in only one or two sectors, versus the deep integration that Facebook and Google have across industries, across the personal and the professional.

There’s been a lot more talk lately of “breaking up Facebook” or “breaking up Amazon,” and that’s a healthy conversation to have. But public provision of the most important things these companies offer is a legitimate alternative that should be discussed alongside antitrust action, deeper regulation, and all the other ways the tech giants might be rendered more pro-social.

A lot of government regulation or public ownership arrives when something about physical space puts limits on competition. A city can’t reasonably have 30 different cable companies running wire through everyone’s front yards, so a regulated monopoly becomes the compromise. It does no one any good to have 8 different TV stations broadcasting on Channel 7, so broadcast licenses are created, regulated, and meted out. The fact that the Internet seems to be boundless — a small-d democratic medium, where 8 or 18 or 8,000,000 broadcasters can send out their signals — led some to decry both regulation and a public role.

We’ve learned better by now — that the Internet’s very structure encourages the kind of dominance we’ve seen concentrated in a few billionaire hands. It’s time for a smart national conversation about it, and I think Zuckerman lays out a strong case that digital public infrastructure is worth consideration.

Photo of the Ambassador Bridge connecting Detroit and Windsor, Ontario — the only privately owned international crossing in the United States — by Alyssa Black used under a Creative Commons license.

Joshua Benton is the senior writer and former director of Nieman Lab. You can reach him via email ( or Twitter DM (@jbenton).
POSTED     Jan. 22, 2020, 10 a.m.
Show tags
Join the 60,000 who get the freshest future-of-journalism news in our daily email.
Postcards and laundromat visits: The Texas Tribune audience team experiments with IRL distribution
As social platforms falter for news, a number of nonprofit outlets are rethinking distribution for impact and in-person engagement.
Radio Ambulante launches its own record label as a home for its podcast’s original music
“So much of podcast music is background, feels like filler sometimes, but with our composers, it never is.”
How uncritical news coverage feeds the AI hype machine
“The coverage tends to be led by industry sources and often takes claims about what the technology can and can’t do, and might be able to do in the future, at face value in ways that contribute to the hype cycle.”