Maybe Facebook Should Abandon the News Feed Altogether.

By devaluing journalism, Facebook risks abandoning the public square. By becoming a true platform, it could *build* that square.

There was a time, about a decade ago, when for a brief moment I thought Facebook would become the platform underpinning the open internet. Such a platform was desperately needed — the open web was a wonderful but messy place that lacked structure and discipline. Facebook’s young CEO claimed he wanted to make the world more “open and connected,” and for a minute it seemed his company might provide the infrastructure to power a people-driven next generation of the web. I even asked him about the idea — a few times, in fact — during interviews on stage at the Web 2 Summit, an event I used to convene back in the early days of the web boom. At the time, he’d point to Facebook’s “Platform” as an initial response to my queries.

A bit of history. Ten years ago, Facebook announced “Platform,” an audacious attempt to corral the World Wide Web’s open, messy nature and funnel developer (and startup) energies into the Facebook universe. The move shook the startup world — most of which were focused on creating independent web sites. But after Facebook’s announcement, every company got busy developing a “Facebook Platform strategy.” Countless hours (and dollars) were expended by firms large and small, all attempting to create their own version of themselves in Facebook’s white and blue walled garden.

As a startup CEO at the time, I resisted Platform — it struck me as madness to build your value on the shifting sands of someone else’s domain unless it had very clear values which aligned with the open web. Facebook Platform’s terms of service were pretty draconian, and I predicted, rightly if ineffectually, that Facebook would end up screwing over its partners as it began to colonize the most profitable platform applications. Not surprisingly, my point of view was mostly ignored. By 2010, white shoe venture firms had launched Facebook application funds, and pretty much everyone thought “social apps” were going to take over the world.

Turns out, Facebook took over the world all by itself. But more on that in a minute.

Platform rolled out as a set of tools and services designed to let other businesses build on Facebook’s core assets. The idea was pretty simple, and utterly in line with how people thought about the world wide web at the time — create APIs that allow others to access your data and services, lay out governing policies, and sit back and watch thousand of flowers bloom.

And bloom those flowers did — thousands upon thousands of Facebook “apps” were built. But most of them were crap — silly games, useless quizzes, the kind of dreck one might expect from a simple system created in the early days of social. (Sheep throwing was a thing. Really). A few got big, and then either got bought or pivoted away from Facebook — Buddy Media comes to mind, or Zynga.

By 2012 Facebook had pretty much abandoned its original Platform vision. It realized that its most valuable asset was the attention of hundreds of millions of consumers, and it optimized its internal platform towards more and more engagement. In so far as the company focuses on third party developers, it’s on those which feed its advertising and data collection services, or which create “content” that drives engagement, and therefore advertising inventory. That sounds terribly cynical, but it’s also true.


But to really understand Facebook’s current dilemma, we must look to News Feed, the “killer app” which allowed Facebook to truly consolidate its power. Launched to great user consternation in late 2006, News Feed came to dominate the Facebook experience within a year or so. As Facebook grew, and began its march to advertising dominance, News Feed became the company’s primary source of data-driven advertising revenue. The company encouraged advertisers and publishers to game the News Feed algorithm so as to gain more followers and engagement. The explosion of mobile, with its feed-driven consumption model, only inflamed the feed’s power. By the time Facebook launched Instant Articles, a “platform” for publishers to deliver content directly into the company’s News Feed advertising machinery, Facebook had become the world’s dominant information-delivery vehicle. Platform was a distant memory. Sure, you could still plug into Facebook’s “platform,” but only if you were helping the company spin Facebook engagement into advertising gold. Publishers didn’t like the deal Facebook offered, and complained that revenue was unpredictable and shallow, but that was life. Facebook had won.

Then Election 2016 happened, and 2017’s news cycle was dominated by how News Feed had contributed to Russian interference, political divisiveness, and filter bubbles. Slowly, Mark Zuckerberg got woke. His new commitment? To fix Facebook.

As 2018 dawns, we’ve seen the first such “fix” — the remarkable pivot of News Feed away from news. Last week the company announced it would tune News Feed toward posts from friends and family, and away from post from brands and publishers. The company had been making noises about doing this for months, but Zuckerberg’s announcement cemented the move. As the news settled in, the media’s core takeaway can best be summarized thusly: Facebook never liked the media, the media caused Facebook all manner of problems (like Fake News and an overweighted responsibility for public discourse), and now, Facebook is done with media. Deal with it.

Perhaps, but here’s what I don’t understand: Why are we all arguing about whether or not Facebook has the capacity to manage media flows on its platform at all? Why can’t we change the narrative, and ask a different set of questions?

Question #1: Why does Facebook have to control everything we see? For whatever reason, we’ve accepted a core premise that “the algorithm” will feed us just the right information. When it doesn’t, Facebook goes to work “fixing” that algorithm, as it just announced last week. But why? Why can’t the consumer take greater charge of his or her own information feed? As I’ve said elsewhere, it’s simply untenable to have one company’s algorithms control the personalized feeds of billions of humans around the world. The only way I see Facebook’s core problem being solved is to open its platform up to third party developers. But of course, that’s super messy. And it breaks Facebook’s current stranglehold on advertising revenues. But….

Question #2: Why not let everyone create News Feed apps? Facebook could curate “base cases” — one for just friends and family, and one for brands and publishers you want to follow. Then anyone, any developer, any kid from Macedonia (OK, bad example), any publisher, any curator like Dave Pell or Jason Hirschhorn could create a feed, and we, the people, get to choose which we consume. Of course there should be advertising in those feeds, and of course Facebook should get a cut. But everyone is used to Apple getting its 30 percent, so why not on Facebook? This of course begs my Big Question….

Question #3: What if Facebook *really* pivoted, and became a true Platform? This is the only way I can see Facebook managing its way out of this mess. Back when we started the Web 2 Summit (2003–4, for those keeping track), my partner Tim O’Reilly penned an essay explaining the vision behind “The Web As A Platform.” It’s worth reading, though it’s dense. To boil it down, we believed that the open web should and would evolve to become a base foundation for services and products that shared key capabilities (data collection/storage/sharing, identity resolution, transaction services, etc). I believed that certain companies would likely find natural monopolies in various aspects of this — Stripe for payments, perhaps, Twilio for telephony integration — but all of these component companies would by necessity be open and integrative. With these interconnected web services comprising the core platform, “applications” on the web could focus on higher level value creation — news curation, branded storefronts, social networking, etc. No one company would try to “boil the ocean” and become the sole source of attention (Facebook), or commerce (Amazon), or navigation and OS (Google/Apple) on the web. The web would be like (much of) the world — open, connected, and driven by individuals making their own choices about what they consume. That was the vision, anyway.

But of course, this isn’t how the web evolved. Instead, those aforementioned three or four major companies effectively privatized huge swathes of the web’s “platform” layer, with Facebook, thanks to its identity data, becoming the most powerful of them all, at least when it came to information and media. And now, well now Facebook is in something of a pickle. It has too much attention — and not of the good kind. It can’t possibly figure out how to best parse that attention for each of us. It’s not fair to even expect it to.


So imagine with me what might happen if Facebook were to become a truly neutral platform — an AWS for attention and identity, if you will. What if the company dedicated itself to a set of stable policies that encouraged other companies to tap into its social graph, its vast identity database, its remarkable engagement machinery?* Instead of squeezing everyone through the monolithic orifice of News Feed, what if Facebook changed the narrative completely, and reshaped itself as a service anyone could tap to create new kinds of value? Facebook could set the rules, take its cut, and watch tens of thousands (millions?!) of applications bloom. More than a few of them, I’d wager, would be extraordinary new interpretations of the News Feed —and because they’d compete in the marketplace of ideas, with individual citizens deciding which of them they’ve decided to consume, Facebook would be off the hook as the sole provider of society’s informational nutrition.

Of course this idea is crazy, complex, fraught, and seemingly impossible. But it sure beats the alternative — where one company, and one company alone, is responsible for determining what information the majority of society consumes. No one — Facebook included — seems to want that anymore.


Facebook’s head of News Feed, Adam Mosseri, will join us on stage at Shift Forum next month.

  • This is the core of the idea I was floating in another recent essay about the creation of a data commons.