The Triumph of Email
Why does one of the world’s most reviled technologies keep winning?
Email, ughhhh. There is too much of it, and the wrong kind of it, from the wrong people. When people aren’t hating their inboxes out loud, they are quietly emailing to say that they’re sorry for replying so late, and for all the typos, and for missing your earlier note, and for forgetting to turn off auto-reply, and for sending this from their mobile device, and for writing too long, and for bothering you at all.
For an activity that’s so mundane, email seems to be infused with an extraordinary amount of dread and guilt. Several studies have linked frequent email-checking with higher levels of anxiety. One study found that constant email-checkers also had heart activity that suggested higher levels of cortisol, a hormone associated with stress—until they were banned from their inboxes.
In the mobile Internet age, checking email is simultaneously a nervous tic and, for many workers, a tether to the office. A person’s email inbox is where forgotten passwords are revived; where mass-mailings are collected; and where pumpkin-pie recipes, toddler photos, and absurd one-liners are shared. The inbox, then, is a place of convergence: for junk, for work, for advertising, and still sometimes for informal, intimate correspondence. Email works just the way it’s supposed to, and better than it used to, but people seem to hate it more than ever.
Over the course of about half a century, email went from being obscure and specialized, to mega-popular and beloved, to derided and barely tolerated. With email’s reputation now cratering, service providers offer tools to help you hit “inbox zero,” while startups promise to kill email altogether. It’s even become fashionable in tech circles to brag about how little a person uses email anymore.
Email wasn’t always like this. We weren’t always like this. What happened?
The computer engineer Raymond Tomlinson sent the first email in 1971. He can’t remember what it said, but people keep asking him anyway. “It was completely ephemeral, so any trace of it is gone,” he said. “There may be a machine that has some memory that was hooked up at the time, but you’d never be able to find it.”
Back then, Tomlinson was developing applications and protocols for the ARPANET, the early network that today’s Internet is based on.* (Today, he’s a principal scientist at BBN Technologies, a research and development arm of the defense giant Raytheon.) In 1971, the idea that anyone other than Tomlinson’s coworkers would want to use email was out of the question. “The computer was not personal,” Tomlinson said. “It was time-shared amongst several dozen users. Most computers were quite expensive—tens if not hundreds of thousands of dollars.”
Email arrived at a time before mobile phones, when it was much harder to reach someone who wasn’t right there with you. “Getting ahold of people, especially those in other time zones, was very difficult,” Tomlinson said. “If they didn't answer the telephone, if you were lucky, maybe they had a secretary—or an answering service if they were really important.”
In building apps for the ARPANET, Tomlinson and his colleagues had talked about some sort of mailbox protocol. One idea was to establish numbered electronic mailboxes so that messages could be printed out then hand-delivered to cubbies with the corresponding numbers. “I looked at that and said, ‘Well, it’s an interesting idea, but it’s way too complicated,’” Tomlinson told me. A simpler method, he thought, would be address messages to individuals. Though the goal was to be able to communicate with engineers working on the ARPANET at other universities, the first email Tomlinson sent was from one computer to another, both standing “literally side by side” in a Cambridge, Massachusetts, lab.
Between the roar of the computers and the whir of the air conditioner required to cool them down, the room was noisy. And the machine Tomlinson used to hit send barely resembled today’s computers. “Brace yourself for a sharp turn,” Tomlinson told me, “There was no monitor.”
Instead, he used a beige terminal the size of a large typewriter, without a mouse or trackpad, for inputting instructions. The terminal itself was something like a Teletype Model 33 KSR, and it was hooked up to a printer that spit out 10 characters per second, all capital letters.
Which means: The first email had to be printed out in order to be read.
Tomlinson’s the one who selected the @ symbol for email addresses, and it stuck—despite a brief period in the 1980s when some service providers experimented with exclamation points and percent signs instead.
In the early days, checking email required a person to log onto a computer and use the keyboard to enter a “type mailbox” command. “The mailbox was just a file and the type command typed the contents of the file onto the paper in the terminal,” Tomlinson said. “Some systems would check the user’s mailbox after they logged in, and if it was not empty, a message like, ‘YOU HAVE MAIL,’ would be printed.” A separate program had to be used to compose outgoing messages, before inbox-outbox functionalities were eventually integrated. “By the end of the 1970s, most of the features of email we take for granted were present,” Tomlinson said.
At first, email was useful, but it wasn’t widely used—it would be decades before electronic mail entered the mainstream. In the 1980s, early adopters flocked to networked services like CompuServe and Prodigy, both of which offered email access, though not necessarily as a central feature. Tim Berners-Lee outlined his idea for the World Wide Web in 1989 at a time when most adults in the United States didn’t own a personal computer. That quickly changed.
By 1995, about one-third of Americans owned computers and 14 percent of them reported having a home Internet connection—mostly sluggish dial-up. As Internet adoption steadily climbed, email became its cultural touchstone, and the inbox became a phenomenon. “If you don’t have an Internet address,” a then-37-year-old New Jersey man told The New York Times in 1994, referring to email, “it marks you as a nobody, as someone who’s over 40. It’s reaching the point that you get socially ostracized.”
America Online, the company that helped millions of Americans explore the web for the first time, was built around the experience of checking mail. Which meant that for millions of people, the experience of going online, from the very beginning, was fundamentally about checking your email. By 1997, electronic mail crept into workplaces and across college campuses. Email became a central plot device in the romantic comedy “You’ve Got Mail” in 1998, and was the subject of the Britney Spears song “Email My Heart” in 1999.
People were in awe of email. They loved it.
Until they didn’t.
The novelty, at some point, faded. Since 1999, Internet use has increased more than tenfold—with the global online population going from about 280 million people to more than 3 billion people, according to Internet Live Stats. Email volume appears to be growing, still, but its share of overall electronic communication has shrunk.
If there’s any clue from the behavior of teenagers as to the direction of a given technology, email appears, well, doomed. Teens barely use it (or Facebook for that matter), opting instead for text messaging and chatting on platforms like Snapchat and Instagram. Three-quarters of teens regularly text one another, according to a 2012 Pew study, while just 6 percent of them exchange emails routinely.
People seem to hate email for the same reasons they once loved it. Email’s underlying triumph, the quality that made it revolutionary, was that you could instantly deliver a written message to someone even if they weren’t there to receive it. (Though fax machines offered some of the same benefits, they were more frequently used for business-to-business communication than person-to-person correspondence.) But leaving messages for people to pick up later means contributing to swelling inboxes that require time to maintain.
Email is neutral, meaning that anyone can email anyone else with an email address. If you have a person’s email address, your message will be delivered no matter who you are—whether the recipient is your oldest friend, your granddaughter, your boss’s boss, or Beyoncé. The year the web was born, this flattening effect was astonishing. Anyone in an organization could communicate directly and immediately with anyone else, “regardless of rank,” as the The New York Times put it in an article about “computer mail” in 1989. That neutrality is part of what makes email so special. It is, however, what makes inboxes overflow, too.
In 2016, instead of being the subject of romantic comedies and love songs, email is at the center of conversations about digital overload and work-life imbalances. The words “drowning,” “avalanche,” and “tyranny” are used. People resent their inboxes because they are not in control of them. Email takes a psychological toll. It “emotionally weighs on us,” said Alex Moore, CEO of Boomerang, which offers a suite of efficiency tools like email scheduling, snooze features, read receipts, and reminders.
“We let email interrupt us dozens and dozens of times a day, and that is awful,” Moore said. “There’s research out there that says every time you get an email notification and you look at it, it takes you 64 seconds to recover. You basically can never work. You’re constantly recovering from the notification.”
“We’re stressing ourselves out,” he added. “We’re living in notification hell. That’s really the thing that’s at the root cause of why people hate email.”
With more communications platforms to choose from, people aren’t using email as they once did. Today, there are too many real-time communications platforms to track. Along with email, people can chat through tweets, Gchat, Yik Yak, Snapchat, Facebook, Instagram, Viber, Skype, HipChat, FireChat, Cryptocat, and—perhaps most popular of all—text messaging.
Slack, a real-time messaging platform built for the mobile era, may be the best known example of what business communications might look like in a post-email world, but many companies bill themselves as inbox destroyers. (It’s not an overstatement to say Slack can vastly reconfigure a person’s relationship with email: The Atlantic has used Slack since 2014, and, for me, it’s been transformative.) In the pre-Slack era, I worked in newsrooms that used Skype and Yammer. Asana is a project-management tool that promises “teamwork without email.” For email-free task collaboration, there’s also Trello and Basecamp, among others.
I’ve sent 58 emails at my job this calendar year, for anyone who scoffs at the whole Slack thing— Casey Kolderup (@ckolderup) December 16, 2015
In Silicon Valley, the question of what comes after email is already dated. In the newsroom where my colleagues and I used Skype, more than five years ago, one colleague, a website developer, refused to use email on principle. Lately, I’ve seen promoted tweets from a company, Ryver, promoting itself as the the product that will replace Slack. In 2011, Robert Half Technology polled 1,400 executives and found that more than half of them believed real-time communications platforms would surpass email by this year. Some people argue that’s already happened.
All this presents an odd paradox. In Internetty circles, especially among people who’ve been on the web for a long time, a common mentality is that open is good. And though email may be despised, it is still a cornerstone of the open web. “Email is the last great unowned technology,” said the Harvard law professor Jonathan Zittrain in an episode of the podcast Codebreaker in November, “and by unowned, I mean there is no CEO of email... it’s just a shared hallucination that works.”
The wonky net-neutrality debate that’s raged for most of this century comes down to this very principle. “It was because the Internet had been designed so cleanly to just allow anything to be built on top of it, and because it was decentralized,” Berners-Lee said in a speech in 2014, “I could just develop the web without asking anybody else permission.”
Stewart Butterfield, the CEO of Slack who likes to remind people he’s been online since 1992, says he isn’t actually out to kill email—even if his company has been celebrated for doing just that. For being the head of a multibillion-dollar tech darling, Butterfield makes his ambitions sound almost parochial. He just wants to have an “improvisational jam session with the whole world,” he said, referring to the inspiration Slack draws from classic early-Internet chat protocols like IRC.
Though Butterfield, too, sees email as “the last great unowned piece of the Internet,” he also says it’s “crazy” that businesses use email for so much intraoffice communication, especially when inboxes are already saturated. “With email, probably 80 percent of it is not coming from a person—and I don't mean just spam,” he said. “Here’s a receipt from your Uber ride, someone’s following you on Twitter, marketing newsletters. And in the business context, there are probably even more messages from systems.”
The idea behind Slack is that, when you’re addressing the same core group of people via email all the time anyway, you might as well have a shared digital space so that people can dip in and out of the conversation as needed. (Just think: No more sifting through email threads in which six people reply-all to say “thanks.”) Slack isn’t about spending less time communicating, Butterfield says, it’s about accomplishing more in that time.
Email’s standing has always depended on an evolving view of what's possible with technology in time and space. Email is no longer tethered to a computer; instead, it’s under most people’s thumbs practically all the time. And that change is driven, almost entirely, by the rise of mobile. In the same way that cellphones made it so a person didn’t have to be at home for them to take a phone call, smartphones have made it so a person doesn’t have to be in front of a computer to respond to an email.
Today, less than a decade since the first iPhone was introduced, more than two-thirds of adults in America have smartphones. People in the United States are increasingly going online exclusively via smartphones, and dropping high-speed Internet connections altogether.
If email represents one kind of “notification hell,” push notifications are the next circle of it. The Internet-everywhere era that smartphones ushered in brought with it vibrating, chiming, knocking, thrumming, anytime alerts from an endless number of people and apps. Direct access to the screen in someone’s pocket is hugely appealing to businesses, whether it’s MyFitnessPal telling you to log your lunchtime calories or The New York Times delivering a breaking-news alert.
Push notifications are the natural extension of email, and with the rise of wearables and Internet-connected-everything, it’s only going to get worse. These notifications act like text messages at a time when people are already texting constantly. A company that wants users to engage with its app can individually target people, and do so in a way that mimics the kind of communication they’re having with their closest friends and family members. Salespeople have also tried to imitate informal, personal communications via email (see also: presidential campaigns, BuzzFeed newsletters, and Louis C.K.), but whereas email is delivered and left to be picked up by the recipient, push notifications are designed to interrupt you at the moment they are sent, and—if your smartphone vibrates—to physically jolt you into knowing that someone wants your attention right now.
“You can get anyone’s attention at the fucking speed of light,” Butterfield said. “The allure of that is so powerful… it’s hard not to check your phone.” That’s part of why Slack last month introduced a “do not disturb” feature, so that people can block off times in which Slack notifications won’t reach them on their phones unless they go looking for them. The flip side of this feature is that anyone can post anything into a Slack channel at any time, without worrying about bothering colleagues at odd hours. In other words, it frames Slack messages as retrievable at the convenience of the person reading them. Which, in a funny way, makes Slack ever so slightly more like email.
Email, it should be noted, works great on mobile. It is an old-school protocol that renders gorgeously on new-school devices. “While the mobile web is a rusting scrapheap of unreadable text, broken advertisements, and janky layouts, normal emails look great on phones!” Alexis Madrigal wrote in 2014. “They are super lightweight, so they download quickly over any kind of connection, and the tools to forward or otherwise deal with them are built expertly and natively into our mobile devices.”
Email’s endurance isn’t just luck. It has improved, too. Spam filters work really, really well. And many providers offer email services that are both free and eminently usable. Gmail will divvy up the marketing from the news headlines from the messages from your brother-in-law. It also recently unveiled a smart auto-reply feature, a time-saver designed to guess how you might want to respond to an email. Early iterations of the service were inappropriately affectionate: When the machine wasn't sure how to sign off, it would default with “I love you,” a detail that’s perhaps sweet enough to make even the steeliest email-haters soften.
Filtering and predictive-response features hint at what email could become in the future, especially as communications continue to splinter off onto other platforms like Slack, Facebook, the forthcoming Google chat app, and text messaging. “Email has had a similar evolution as snail mail,” said Michael Heyward, the CEO of Whisper, a social network where people can communicate anonymously. “Both started off as a primary means of communication that people were excited about, and now, you mainly see spam—bills, marketing promotions—and occasionally, an important piece of information will come through.”
So there’s incentive for service providers to make receiving email more efficient—not just sorting out the junk messages, but using machine learning to determine which messages are highest priority. Not that it’s an easy task. Hundreds of billions of emails are sent each day, amounting to some 75 trillion emails per year. Three years from now, that number is expected to go up to 90 trillion annually, according to several estimates.
White-collar workers check their inboxes an average of 77 times a day, according to research by Gloria Mark, an informatics professor at the University of California, Irvine. (If that sounds low to you, she found some workers check email far more frequently, up to 343 times a day or more.) The more time people spend focused on email, Mark has found, the less happy and productive they are.
“Email has evolved into a weird medium of communication where the best thing you can do is destroy it quickly, as if every email were a rabid bat attacking your face,” Paul Ford wrote last year. “Yet even the tragically email-burdened still have a weird love for this particular rabid, face-attacking bat.”
That love may not be all that weird, though—especially as email’s competitors, with push notifications, become more annoying. Email works. It’s open. It’s lovely on mobile. And as other forms of communication theoretically lighten the burden email places on people, perhaps it will become more tolerable again. The guilt people often associate with email is, after all, not technological. (Remember, telephone answering machines produced a similar wave of “paranoia and guilt” when the devices were new, according to a 1979 New York Times article.) “That has to be a human feature,” said Tomlinson, the man who sent the first email. “Email does not produce guilt.”
“It may be called something else, it may be embedded within some other app. We may even abandon the protocols. But I don’t think it's going away,” he said. “Email is always going to have a place.”
* As with many key moments in technological history, the question of when and how the first email was sent has been disputed. You can read more about some of that controversy here, here, here, here, and here.