Tristan Harris’ first big idea for the tech industry, the Time Well Spent movement, was an outsized success. Today, he unveiled the sequel — a kind of unified theory of how tech platforms are undermining humanity. His idea, which he calls “downgrading,” attempts to explain everything from smartphone addiction to political polarization. Is his diagnosis correct? And, if so, what’s the solution?
First, some relevant context. Six years ago, Harris was a product manager at Google who published a presentation for his fellow employees to read. Observing how often Google services compelled people to check their email and smartphone notifications, Harris called on his fellow employees to build systems that gave people time back. “Change like this can only happen top-down, from large institutions that define the standards for millions of people,” Harris wrote. “And we’re in a great position to do something about all this.”
The presentation spread quickly inside Google. But despite an initial rush of enthusiasm, Harris became convinced that he could be more effective working on these issues outside the company. He gave a TED talk about how tech companies could protect us from distractions, and formed the Center for Humane Technology with some friends to lobby them to do better.
Along the way, Harris began using the phrase “time well spent” to describe his goal. He wasn’t asking us to abandon our phones — only to use them intentionally, and with care.
Harris was not alone in calling for tech companies to build time- and attention-management features. But I believe his coinage of the phrase “time well spent” helped to catalyze, and accelerate, the movement. In three little words, Harris conveyed a big idea, and made it easier for product managers at big companies to discuss in shorthand. Harris and his colleagues pushed the idea forward in interviews and essays, and in private conversations with employees at the companies where they hoped to effect change.
Last January, I wrote that “time well spent” was shaping up to be tech’s next big debate — and six months later, it was effectively over, as Apple, Google, and Facebook had all added features designed to help users measure their time using those companies’ products and to manage their usage.
All of which is to say that Harris took the stage in San Francisco today from a position of strength. The center he co-founded had gathered a couple hundred people, including current and former employees of Google, Facebook, and Twitter, and walked us all through his vision of “downgrading.”
Tech companies, he argues, have “downgraded” humanity by promoting shortened attention spans, outrage-fueled dialogue, smartphone addiction, vanity, and a polarized electorate. Harris called for tech companies to enable a new “race to the top,” centered on building tools to help people focus, find common ground, promote healthy childhoods, and bolster our democracy.
Few would disagree with Harris’ aims. Much less clear, though, is how he intends to get there. Even if you accept that all of the societal ills Harris names have a single common cause — and it seems like a reach to me — it will take many different tools to solve them. And what those tools might be, Harris didn’t say.
Instead, he promised three things from his organization in the coming months. One, a guide for product organizations devoted to promote more humane designs. Two, a podcast about these issues, called Your Undivided Attention, coming June 10th. And finally, a full-fledged conference will arrive in 2020. As of today, the Center for Humane Tech is as much a (nonprofit) media company as it is a movement.
Still, Harris speaks with passion about what he sees as a crisis.
”This is a civilizational moment in a way I’m not sure we’re all reckoning with,” Harris said on stage. “It’s a historical moment when a species that is intelligent builds technology that … can simulate a puppet version of its creator, and the puppet can control the master. That’s an unprecedented situation to be in. That could be the end of human agency, when you can perfectly simulate not just the strengths of people but their weaknesses.”
Harris’ earnestness has resulted in mocking in some quarters. Wired’s Nick Thompson, who got a preview of Harris’ speech today, quoted an unnamed tech executive at one of the big platforms saying this:
“Tristan sees humans as pawns incapable of managing their own lives. He thinks designers are infinitely powerful and can coerce people to do whatever they want. It is a pure farce.” The executive adds, “I like to imagine Tristan reviewing the latest restaurant. ‘They have clearly intentionally added flavor to this dish to make me want to come back and visit this business again. What scoundrels!’”
Meanwhile, one attendee called today “the most offensive event I have ever been to on many many levels,” saying it centered the voices of designers and engineers, and argued that Harris’ approach to reversing “downgrading” borrows all the frameworks that got us to this place.
Still, Harris speaks the Silicon Valley dialect for a reason: his target audience is the people already working at the big platforms, who remain in the best position to make radical change. Perhaps the next generation of entrepreneurs will adopt human-centered design principles and make today’s giants irrelevant. But until they do — and in case they don’t — Harris is keeping the pressure squarely on them. It’s impossible to say whether his approach is likely to work again — but it didn’t seem particularly likely the first time, either.
Fresh off of passing the General Data Protection Regulation to discourage companies from collecting too much data about its citizens, the European Union is … creating a unified database of all its citizens along with biometrics. OK! From Catalin Cimpanu:
Per its design, CIR will aggregate both identity records (names, dates of birth, passport numbers, and other identification details) and biometrics (fingerprints and facial scans), and make its data available to all border and law enforcement authorities.
Its primary role will be to simplify the jobs of EU border and law enforcement officers who will be able to search a unified system much faster, rather than search through separate databases individually.
Ben Collins reports on a pro-Trump Twitter botnet that was unearthed over the weekend:
A network of more than 5,000 pro-Trump Twitter bots railed against the “Russiagate hoax” shortly after the release of special counsel Robert Mueller’s report last week, according to data gathered by a prominent disinformation researcher and analyzed by NBC News. The network illustrates the ongoing challenge Twitter faces in persistent efforts to manipulate its platform.
These bots, however, did not appear to come from Russia. Instead, the bots had ties to a social media operation that previously pushed messages backing the government of Saudi Arabia and were connected to a person who claimed to be a private social media consultant, according to internet domain and account registration records. The bots, which were created last November and December, were pulled down by Twitter on Sunday night for breaking the social network’s rules against “manipulation,” the company said.
And you’ll never believe what they talked about! Here’s Tony Romm with a story that left me staring into the middle distance for a long time today:
A significant portion of the meeting focused on Trump’s concerns that Twitter quietly, and deliberately, had removed some of his followers, according to a person with direct knowledge of the conversation who requested anonymity because it was private. Trump said he had heard from fellow conservatives who had lost followers as well.
But Twitter long has explained that follower figures fluctuate as the company takes action to remove fraudulent spam accounts. In the meeting, Dorsey stressed that point, noting even he had lost followers as part of Twitter’s work to enforce its policies, according to the source.
Makena Kelly and Adi Robertson explore how Sri Lanka’s temporary ban on social media could lead to worrisome speech restrictions down the road:
Concerns over online misinformation have already made it into law in many countries. Sri Lanka isn’t the only country to block Facebook during a crisis. Facebook’s transparency report shows that other countries like Cameroon, Indonesia, and Iran have seen internet disruptions as well, according to a Facebook transparency report. France, Singapore, and Russia have passed laws designed to curb “fake news” on social media. The UK recently proposed a fine on internet platforms that don’t remove harmful content. And after a mass shooting in Christchurch, New Zealand, the government blocked a handful of sites that hosted video of the attack.
Danny O’Brien, international director of the Electronic Frontier Foundation, acknowledges that propaganda and misinformation is amplified by social media’s vast scale. “In terms of the sheer numbers of people involved in that, then it’s definitely a sort of quantitative difference,” he says. At the same time, he notes that hatemongers have always taken advantage of mass communication systems, and repressive governments don’t draw a line between harmful misinformation and legitimate criticism.
April Glaser reports that right-wing extremists have been raising money on Facebook:
After the New York Times reported on the group’s activities last Thursday, PayPal and GoFundMe stopped servicing them. Still, there’s evidence that similar extremist militia groups have used Facebook itself as a donation platform—the Mountain Minutemen’s leader, Robert Crooks, was able to fundraise $1,600 on the site just last month. Another armed militia group, the Three Percent United Patriots, maintains a Facebook group with more than 19,700 followers and still solicits donations via PayPal. Even as fundraising sites stop servicing them, Facebook offers a centralized way for supporters to communicate in private messages. Those who wish to support the United Constitutional Patriots can easily send a note to ask how to best send funds.
Maggie Fick and Paresh Dave identify “31 widely spoken languages” that still have no official support on Facebook, posing challenges for moderators:
Similar issues abound in African nations such as Ethiopia, where deadly ethnic clashes among a population of 107 million have been accompanied by ugly Facebook content. Much of it is in Amharic, a language supported by Facebook. But Amharic users looking up rules get them in English.
At least 652 million people worldwide speak languages supported by Facebook but where rules are not translated, according to data from language encyclopedia Ethnologue. Another 230 million or more speak one of the 31 languages that do not have official support.
Google employees who helped organize a November walkout of thousands of employees say they’ve been demoted and told to drop their concerns about ethics, Nitasha Tiku reports.
Shannon Liao has news of a Microsoft employee protest:
Microsoft employees have put forward a petition in defense of a trending GitHub repository they believe could be under threat of Chinese censorship. The repository, called 996.ICU, was established in late March by Chinese tech workers who were protesting extreme overwork. The stories name some of the biggest companies in China, including Alibaba, Huawei, ByteDance, DJI, Tencent, Vivo, and others.
The protest has enraged Chinese censors, and many local browsers are already blocking access to 996.ICU, including browsers from Tencent, Alibaba, Xiaomi, and Qihoo 360. Still, the repository remains available to Chinese users under alternate browsers, and GitHub has not moved to take it down.
Rachelle Hampton writes about the Twitter users who were posting early warnings of Twitter’s partial takeover by bots meant to inflame societal tensions in the United States:
Exposing #EndFathersDay ultimately took the work of a group of black women who were determined not to let the ruse spiral, sensing just how poisonous this kind of trolling could be. And yet, in the years since, even as journalists have publicly asked themselves how they missed the rising threat posed by far-right extremists radicalized online, somehow one of the earliest crowdsourced anti-misinformation campaigns on the internet has been mostly ignored by the mainstream media. To I’Nasah Crockett, who, along with Hudson, helped uncover the #EndFathersDay hoax, watching the events of the past few years has made her feel like she was “a canary in a coal mine.”
Twitter had a good quarter. Monthly users were down, but daily users were up. (The metric “mDAU” here refers to “monetizable daily active users,” which is a bad name for a metric that Twitter made up.)
Twitter reported 134 million average mDAUs for the first quarter, compared with 120 million a year earlier. In the fourth quarter, Twitter said it had 126 million mDAUs.
In the U.S., Twitter reported 28 million average mDAUs forthe first quarter, compared with 26 million a year earlier. It reported 105 million average international mDAUs for the first quarter, compared with 94 million a year earlier.
Snap also had a good quarter. Especially if you ignore the fact that it lost $123 million:
Snap is heading in the right direction again. The company revealed in its earnings release today that its daily user base has grown by 4 million people globally. It now has 190 million daily active users, up from the 186 million people who had consistently been using the platform for the last two quarters. This updated number is still 1 million people short of Snapchat’s peak user base since it went public in 2017, but this is still good news for Snap.
Snapchat CEO Evan Spiegel said in prepared remarks that the platform reaches more 13- to 34-year-olds in the US than Instagram, but didn’t elaborate on why or how its user base suddenly grew. He says Snapchat reaches 75 percent of 13- to 34-year-olds and 90 percent of 13- to 24-year-olds.
Apps designed to help people quit smoking and fight depression are sharing large amounts of data with third parties, including the big tech platforms. Rachel Siegel reports:
That’s according to a study published Friday in the journal JAMA Network Open. Researchers say the findings are especially important in mental health, given the social stigmas and the risks of having sensitive information shared unknowingly. And because many health apps aren’t subject to government regulation, researchers say, consumers and clinicians must contend with what information is being entered into these apps — and who else can access it.
“Digital data doesn’t go away,” said John Torous, a co-author of the report. “A part of the risk is that we don’t fully know who is going to put this data together, when and where it’s going to show up again and in what context. … Data seems to end up in the hands of the wrong people more and more.“
Kerry Flynn reports that publishers have begun some light experimentation on TikTok. Weirdly, none of them has yet uploaded a documentary on the 1989 Tiananmen Square protests!
As of February, NBC News’ “Stay Tuned” has posted 26 videos on TikTok. The 15-second videos feature one of the show’s three hosts — Savannah Sellers, Gadi Schwartz and Lawrence Jackson — commentating on news like the season premiere of “Game of Thrones” or participating in challenges like pineapple pulling.
Last month, ESPN joined TikTok with a video set to TikTok-sensation-turned-billboard-hit “Old Town Road” featuring ESPN sports analyst Stephen A. Smith in a suit and then a cowboy hat. ESPN’s next video is a montage of basketball players getting hit by imaginary Pokeballs. The network’s fifth and most recent video is a back-and-forth staring contest between Bran of “Game of Thrones” and basketball player Draymond Green.
Taylor Lorenz reports that cool teens are no longer trying to look cool in their Instagram pictures:
In fact, many teens are going out of their way to make their photos look worse. Huji Cam, which make your images look as if they were taken with an old-school throwaway camera, has been downloaded more than 16 million times. “Adding grain to your photos is a big thing now,” says Sonia Uppal, a 20-year-old college student. “People are trying to seem candid. People post a lot of mirror selfies and photos of them lounging around.”
Take Reese Blutstein, a 22-year-old influencer who has amassed more than 238,000 followers in just over a year by posting unfiltered, low-production photos of herself in quirky outfits. (A recent flash photo into a mirror with her dog picked up more than 5,000 likes). She, like many members of her generation, doesn’t stress about posting almost the exact same photo twice in a row, something first-generation influencers wouldn’t dream of. “I’m not afraid to over-post. I don’t think, Oh, will this mess up how my feed looks,” she says. “I don’t think too much about it. If I like an image, I just post it.”
Janko Roettgers reports on some nice incremental upgrades to Reddit:
After testing them with “Game of Thrones” fans and a few dozen other Subreddits for the past couple of months, Reddit is now launching two new post types to its entire user base: Events will help communities with discussions about TV show episodes, awards shows, breaking news and more; Collections will help to curate posts, and make Reddit’s fast-paced discussions a bit easier to digest for newcomers.
Events posts help moderators of a Reddit community to schedule posts related to an upcoming event, and then facilitate timely discussions as that event unfolds.
Zeynep Tufekci argues that tech platforms’ prediction algorithms have become so good that there is now no meaningful opt-out available when it comes to privacy features:
Because of technological advances and the sheer amount of data now available about billions of other people, discretion no longer suffices to protect your privacy. Computer algorithms and network analyses can now infer, with a sufficiently high degree of accuracy, a wide range of things about you that you may have never disclosed, including your moods, your political beliefs, your sexual orientation and your health.
There is no longer such a thing as individually “opting out” of our privacy-compromised world.
Lux Alptraum argues that Instagram’s new guidelines for “sexually suggestive” content are at high risk of being sexist in their implementation:
There are already some signs that Instagram’s moderation algorithm is going to be shaped by a biased idea of indecency—one that sets a far lower bar for women than for men. TechCrunch obtained sample photos about the moderation effort, including two images that could be considered sexually suggestive. One featured a fully clothed, headless man grabbing at his crotch; the other, a woman in lingerie seated on a bed. You could interpret this to mean that men are deemed to be sexually suggestive by virtue of their actions, while women gain that status simply by having a body that someone else deems sexually attractive.
And finally …
Jack Dorsey is famous for wearing beanies on stage. Beloved Twitter character Darth had some fun with that idea in this take on Dorsey’s meeting with the president today.
Talk to me
Send me tips, comments, questions, are your ideas for humane tech: firstname.lastname@example.org.