by Biz Stone
Jack ate some soup. Then he said, “I’m going to be like Steve Jobs. I’m going to come back one day.” When he said that, his posture changed, as if remembering that Steve Jobs had also been pushed out of his own company made today easier to bear.
As I’d done at SXSW, I wrote Jack a short, classy speech (which included a whole section about how I was charming, funny, and good-looking). This one praised the team and was generally positive, even though he was feeling not so positive.
The truth is that the ghost of Ev as CEO had been there all along. Most of the team had come over from Odeo, where he’d been our CEO. In my weekly internal emails, I often referred to Jack as our “fearless leader,” trying to bolster his image. Now I needed to steer the ship back toward Evan.
After he left, Jack and I noodled over an idea for an iPhone app that would help people keep a diary. We met at night at wine bars and worked on it for fun, to fire up our brains, and to have a reason to continue working together. Then Jack disappeared for two weeks. When he came back he told me he was working on a new project with a guy named Jim.
“Turns out using the headphone jack in a smartphone, you can read the magnetic strip of a credit card. So you can turn a phone into a credit card reader,” he told me.
“Whoa, that’s crazy,” I said.
That idea was Jack’s new startup, Square. I became an angel investor. I knew I wanted to have some level of participation in any project Jack was spearheading.
With the shift in leadership and the tech issues, the team was fractured. We were a laughingstock in the tech world. Programmers were blaming each other. As always, when all else fails, turn to Star Trek. There is a Next Generation episode called “Attached.” It focuses mainly on Captain Picard and Dr. Crusher. In it they’re alone on the planet Kesprytt III. Some of its inhabitants capture them and implant them with “transceivers” that allow them to hear each other’s thoughts. At one point, while they’re walking, they lose their way. Captain Picard says, “This way.” But the doctor, reading his mind, says, “You don’t really know, do you?” The captain admits that sometimes being a leader means cultivating the appearance of confidence.
That’s my leadership move. Saying “This is what we’re going to do. This is the right thing,” gives everyone the sense of a common mission. We needed to focus on something that felt bigger than our off-balance company. The 2008 presidential election was on the horizon, and both candidates had Twitter accounts. Election Night was going to be important for Twitter. Yes, it was a historic election, and sure, the results would determine the course of our country. But all I was focused on was (a) whether Twitter would stay up and running and (b) how we could use this big event to resurrect our team morale. To make us feel team-y again.
In the months leading up to the election, we all worked hard, as a team, to fix the capacity problems that had plagued us.
The candidates may have thought it was their show, but like a news team, we felt it was ours. And I have to say, it worked. The team came together.
The week of the election was a big week for Twitter, not to mention for the United States, and the world. I sent the team a rallying email with the header: “Adding a New Feature to Democratization of Information.” It read:
Folks,
Birds in flight have an amazing ability to move as one—immediate feedback and simple rules create something gracefully fluid and seemingly choreographed. In the spring of 2007, we caught a glimpse of people harnessing a similar power for a new kind communication. South by Southwest 2007 introduced us to the incredible relevance of Twitter during a shared event.
Now the world is watching as one of the most massively shared events in US history unfolds before us. Twitter is positioned to support this election process like nothing else before—37 members of Congress are Twittering, both candidates have active accounts, millions of citizens are reacting to issues in real time, and political activists are organizing protests—all using Twitter, all moving as one.
Okay, it was a little over the top. But when I wrote that note to the Twitter team, I wanted them to come away from it thinking, Holy shit, my work is important! I wanted them to share the email with their husbands or wives, saying, Look! What I’m doing matters.
A crew of us stayed late on Tuesday, Election Night, to make sure everything was running smoothly. We invited about fifty people to join us for food and drinks, and watched the election results on the big screen. What happened next was amazing. Twitter exceeded and sustained normal capacity by 500 percent—without breaking a sweat.
The servers stayed up. And we had our first African American president! That was the order of the news in our offices that night. Call us Ishmael—our whale was, for the moment, nowhere to be seen. And with the great beast safely out of sight, we welcomed the first U.S. president to have an official Twitter account.
A few weeks later, there was a series of terrorist attacks in Mumbai, India. People in the middle of the crisis used Twitter to report what was happening in real time, and in some cases Twitter served as a lifeline.
People everywhere were finding the reasons Twitter was relevant in their lives—from reading movie reviews to helping the homeless to spontaneously raising money for world causes. While we were buckled down, focusing on performance, the rest of the world was figuring out what Twitter was for.
I wasn’t the only person having big ideas about how Twitter could be used. On July 30, 2008, a 5.4-magnitude earthquake hit Southern California. The official time of the quake was 11:42 a.m., but the Tweets announcing it came in earlier than that. Nine minutes later, at 11:51, the Associated Press sent out a fifty-seven-word alert on its wire service. In those nine minutes, Twitter saw thirty-six hundred Tweets with the word quake in them. In that nine-minute news gap, we had assembled an entire book’s worth of firsthand accounts of the event. Of course Twitter isn’t a traditional news service. Our reports don’t disseminate reliable data and facts. They’re user generated and 140 characters or less. What Twitter has to offer is speed. The AP is as fast as it can be. But Twitter has a global user base sending messages every second, in real time. Whether or not it’s the future of news, it’s at least complementary. Getting information quickly is one of the best things you can do on the planet. Twitter instantly connects us to what is happening in the world.
In the Bay Area, everything always comes back to earthquakes and how best to endure them. That July 30, people were tweeting as they were experiencing the earthquake; they couldn’t resist. The impulse to tweet, mid-quake, was too strong. Those Tweets created a map measuring the impact and reach of the event. And the Tweets spread faster than the quake. Twitter could beat earthquakes to the punch.
Wait a minute. This thing isn’t just for breakfast anymore. It’s not just for reporting what’s happened. It’s able to predict what’s gonna happen. Experts in the emergency response field saw immediately that Twitter could do something that their systems couldn’t. They started calling us, wanting to work with us to make Twitter an official part of the emergency response service, but I told them it was too early. Our service wasn’t reliable enough yet. We didn’t want someone to die because Twitter had gone down.
Nonetheless, after that earthquake, we, and the world, saw Twitter’s potential in an even greater light than before. In order to realize that potential, we needed to be both ubiquitous and reliable. We were charging ahead on both fronts.
In January of 2009 we were in our Bryant Street offices, the place we’d moved to after the funky office on South Park, and were in the middle of what we call Tea Time, a weekly all-hands meeting. Tea Time is a take on Google’s tradition of TGIF, in which every Friday they had free beer and snacks. Jack likes tea, so he thought that, instead, we could meet, talk about the week’s achievements, and enjoy tea and crackers together. But as soon as people realized there was beer in the fridge, there was no turning back. That particular day, we were hosting a Wired reporter, who was spending the week with us for an article ab
out Twitter. He was sitting discreetly in the corner, not looking to report the specifics of the meeting but getting a general sense of what it was like at Twitter. Then he spoke up.
“Hey guys, I don’t want to interrupt, but a plane just crash-landed in the Hudson River. A guy on the ferry that’s helping with the rescue took a picture with his iPhone and tweeted it.”
The meeting broke up. We all went over to look at his computer. It was a perfect picture, showing people in business suits standing on the wing of a US Airways plane in the middle of the Hudson.
The 2008 election had been the definitive moment for us because we’d pushed the technology to a new capacity. But this was a definitive moment for the role of Twitter in the realm of messaging. We all have many different options for contacting people electronically: email, text, IMs, Tweets. There is a time and a place for each. When a plane lands in the Hudson in front of you, that’s a Tweet. That’s the ultimate Tweet. You don’t email a friend that. You tweet it.
On April 7, 2009, I came into work to find my in-box full. My office phone was loaded with messages. It was the press, and they were all asking the same question: What was Twitter’s role in today’s student riots in Moldova?
Um . . . where?
I wanted to write back saying, “Well, we didn’t like what was going on in Moldova, so we pushed the big red Moldova button here on our wall at Twitter and triggered the revolt.”
Instead, I looked up Moldova on Wikipedia. It turned out that students in the country between Romania and Ukraine—I knew those two—had organized to protest Moldova’s preliminary election results, suspecting that they were fraudulent. Because they had used Twitter to organize, the media were calling it the Twitter Revolution.
The user stories I’d envisioned were all coming true, and more so. All of this from an app that began with Jack and me sharing what we were having for breakfast. I no longer had to tell our employees that what they were doing was important. It was clear.
As I explained in an article in the Atlantic on October 19, 2010, there was of course some backlash against the idea of Twitter changing the world. In the New Yorker, Malcolm Gladwell wrote, “Some of this grandiosity is to be expected. Innovators tend to be solipsists. They often want to cram every stray fact and experience into their new model.” This bugged me because we weren’t the ones taking credit for the Moldova protests. On the contrary, we’d been going out of our way to make it clear that we didn’t think Twitter was the voice of revolution. Twitter was just a tool that people were using to do great things. And wasn’t that amazing enough? If you give them the right tools, people do great things. Nobody said that the telephone brought down the Berlin Wall, but were phone calls made? Hell yeah! Twitter was proof that leaderless self-organizing systems could be true agents of change.
As the Arab Spring began at the end of 2010, it became even more important for me to clarify Twitter’s role.
Activists in Arab countries used Twitter and other services like Facebook to organize uprisings. It got so that we could practically predict the next revolution. We’d start to see mounting Tweets in a certain area and we could have made a phone call, “Hey, dictator, you might want to flee.”
Suddenly, as the Arab Spring progressed, every major news outlet wanted me to come on and talk about what was happening. My instinct was not to do that. Not just because I was afraid of being outed as an idiot when it came to global affairs, but I felt that it wasn’t right to gloat—or even focus on what all this meant for our business. People were dying. I wasn’t going to go on TV and say, “Yeah, look at us! We’re a great company!”
While we were glad to be a visible part of the changes taking place, I wanted to be very careful about what our role was. We didn’t have a publicist or anything, so I was the de facto guy deciding if we’d talk to members of the press and about what. So I decided not to talk to any of them. Some of our board members and closest investors were kinda like, “What? Are you crazy? This is huge international news exposure.” They had a point—whenever we went on TV, a million new people signed up to Twitter—but I still wanted to say no to all the major media outlets. I didn’t want to piss them off. I hoped they would talk about Twitter at some point; I just couldn’t endorse it under those circumstances. So I wrote to Raymond Nasr, my friend and communications adviser. I forwarded him the email I wanted to send in response to the media requests. It was short, and it basically said, “Thank you for your interest, but we’re not talking about this.” Raymond, who has always had a great economy of language, said, “It’s perfect. I’d just add the word inappropriate.”
So I sent out an email saying, “Thanks so much for the opportunity, but we don’t think it would be appropriate for us to do the interview or to offer comments beyond what we’ve already addressed to the public on our company blog.”
Most of the responses I received read, “Understood.”
When I wanted to get the job at Blogger, I’d visualized myself working there. I believed that kind of visualization had the power to make things happen. Now, as the user cases I’d envisioned for Twitter all came true, it felt like a waking dream. Things had gotten very serious very quickly. Suddenly we had to make choices about how we interacted with governments.
Occasionally we needed to take down the service for maintenance. Whenever we did that, we put up a note warning the users. But in June 2009, when we put up our customary maintenance notice, we immediately got about a hundred calls and emails saying, “You can’t take the service down then! There’s a scheduled demonstration in Iran.” The Iranian government had shut down other means of communication, and Twitter was considered vital.
Among all the emails we received during that incident, one message stood out. It came from one of our board members. A U.S. government employee had sent a note to him, and he forwarded it as an FYI.
The State Department didn’t want Twitter to go down for maintenance.
Jason Goldman and I hashed out the decision. We needed to do this maintenance—we’d already postponed it thirteen times, and if we didn’t do it soon, the system would possibly break forever.
Finally I said, “Let’s move the maintenance one more time.” It wasn’t because I wanted to follow the State Department’s orders—I didn’t pretend to understand the situation—but because Twitter was supposed to work, and our job was to keep it up and running. What was happening in Iran was directly tied to the growing global significance of Twitter and its importance as a communication and information network.
So we rescheduled the planned maintenance for midafternoon, which was the middle of the night in Iran. To an outside observer, it may have looked as if the State Department had made a call on their red phone and we scrambled to make the change. The thinking was: If we postponed maintenance because the U.S. government had asked us to, then what else were we supposed do for them? But the government did not have decision-making power at Twitter. We didn’t want to help our government, or any other government. We had to stay a neutral technology provider.
On June 16, 2009, the day after the maintenance, I posted the following message to the Twitter blog:
Twitter is back and our network capacity is now significantly increased. The planned maintenance that we moved from last night to this afternoon was a success and it took half the time we expected.
When we worked with our network provider yesterday to reschedule this planned maintenance, we did so because events in Iran were tied directly to the growing significance of Twitter as an important communication and information network. Although presumed impossible if not extremely difficult, we decided together to move the date. It made sense for Twitter and for NTT America to keep services active during this highly visible global event.
It’s humbling to think that our two-year-old company could be playing such a globally meaningful role that state officials find their way toward highlighting our significance. However, it’s important to note that the State Department does not have access to our deci
sion making process. Nevertheless, we can both agree that the open exchange of information is a positive force in the world.
Our stance was that we were government-neutral. We were a communications tool. We didn’t help any state, whether for the sake of revolution or the aid of government investigations. I was proud when, exactly four years later, on June 7, 2013, Claire Cain Miller wrote an article in the New York Times titled, TECH COMPANIES CONCEDE TO SURVEILLANCE PROGRAM, about PRISM—the U.S. government’s secret surveillance program—and said that when the National Security Agency came to Silicon Valley companies asking for user data, “Twitter declined to make it easier for the government.”
Alexander Macgillivray, our general counsel—whom everyone called Amac—had done everything he could to legally support our stance. We weren’t in service to any government, and we made any government attempts to access our user information a huge pain in the ass.
We tried to keep our goal pure: to connect people everywhere instantly to what was most meaningful to them. For this to happen, freedom of expression was essential. Some Tweets might facilitate positive change in a repressed country, some might make us laugh, some might make us think, some might downright anger a vast majority of users. We didn’t always agree with the things people chose to tweet, but we kept the information flowing irrespective of any view we might have about the content.
We believed that the open exchange of information would have a positive global impact. This was both a practical and an ethical belief. On a practical level, we simply couldn’t review all one-hundred-million-plus Tweets created and subsequently delivered every day. From an ethical perspective, almost every country in the world is in agreement that freedom of expression is a human right.