2006–2007: Creation and initial reaction

A sketch, c. 2006, by Jack Dorsey, envisioning an SMS-based social network

"Twitter all began with status-sharing service TXTmob," explains an article on TNW.^[36] Tad Hirsch, a student and activist associated with the Ruckus Society, the Institute for Applied Autonomy, and later the MIT Media Lab, built the basic first application to help activists organize protests at the 2004 Republican National Convention.^{[37][36][38][39]} Inspired by TXTmob and other SMS sharing applications of the day, members of the podcasting company Odeo had an "all-day brainstorming session" to decide on building a new application. Jack Dorsey, then an undergraduate student, claims to have introduced the group to the idea of an individual using an SMS service to communicate with a small group.^[40] The original project code name for the service was twttr, an idea that Williams later ascribed to Noah Glass,^[41] inspired by Flickr and the five-character length of American SMS short codes. The decision was also partly due to the fact that the domain twitter.com was already in use, and it was six months after the launch of twttr that the crew purchased the domain and changed the name of the service to Twitter.^[42] The developers initially considered "10958" as the service's short code for SMS text messaging, but later changed it to "40404" for "ease of use and memorability".^[43] Work on the project started in February 2006.^[44] Dorsey published the first Twitter message on March 21, 2006, at 12:50 p.m. PST (UTC−08:00): "just setting up my twttr".^[45] Dorsey has explained the origin of the "Twitter" title:^[46]

...we came across the word "twitter", and it was just perfect. The definition was "a short burst of inconsequential information", and "chirps from birds". And that's exactly what the product was.

The first Twitter prototype, developed by Dorsey and contractor Florian Weber, was used as an internal service for Odeo employees.^[44] The full version was introduced publicly on July 15, 2006.^[47] In October 2006, Biz Stone, Evan Williams, Dorsey, and other members of Odeo formed Obvious Corporation and acquired Odeo, together with its assets—including Odeo.com and Twitter.com—from the investors and shareholders.^[48] Williams fired Glass, who was silent about his part in Twitter's startup until 2011.^[49] Twitter spun off into its own company in April 2007.^[50] Williams provided insight into the ambiguity that defined this early period in a 2013 interview:^[51]

With Twitter, it wasn't clear what it was. They called it a social network, they called it microblogging, but it was hard to define, because it didn't replace anything. There was this path of discovery with something like that, where over time you figure out what it is. Twitter actually changed from what we thought it was in the beginning, which we described as status updates and a social utility. It is that, in part, but the insight we eventually came to was Twitter was really more of an information network than it is a social network.

2007–2010

The tipping point for Twitter's popularity was the 2007 South by Southwest Interactive (SXSWi) conference. During the event, Twitter usage increased from 20,000 tweets per day to 60,000.^[52] "The Twitter people cleverly placed two 60-inch plasma screens in the conference hallways, exclusively streaming Twitter messages," remarked Newsweek's Steven Levy. "Hundreds of conference-goers kept tabs on each other via constant twitters. Panelists and speakers mentioned the service, and the bloggers in attendance touted it."^[53] Reaction at the conference was highly positive.^[54] Twitter staff received the festival's Web Award prize with the remark "we'd like to thank you in 140 characters or less. And we just did!"^[55]

The company experienced rapid initial growth. In 2009, Twitter won the "Breakout of the Year" Webby Award.^[56][57] On November 29, 2009, Twitter was named the Word of the Year by the Global Language Monitor, declaring it "a new form of social interaction".^[58] In February 2010, Twitter users were sending 50 million tweets per day.^[59] By March 2010, the company recorded over 70,000 registered applications.^[60] As of June 2010, about 65 million tweets were posted each day, equaling about 750 tweets sent each second, according to Twitter.^[61] As of March 2011, that was about 140 million tweets posted daily.^[62] As noted on Compete.com, Twitter moved up to the third-highest-ranking social networking site in January 2009 from its previous rank of twenty-second.^[63]

Jack Dorsey, co-founder and former CEO of Twitter, in 2009

Twitter's usage spikes during prominent events. For example, a record was set during the 2010 FIFA World Cup when fans wrote 2,940 tweets per second in the thirty-second period after Japan scored against Cameroon on June 14, 2010. The record was broken again when 3,085 tweets per second were posted after the Los Angeles Lakers' victory in the 2010 NBA Finals on June 17, 2010,^[64] and then again at the close of Japan's victory over Denmark in the World Cup when users published 3,283 tweets per second.^[65] The record was set again during the 2011 FIFA Women's World Cup Final between Japan and the United States, when 7,196 tweets per second were published.^[66] When American singer Michael Jackson died on June 25, 2009, Twitter servers crashed after users were updating their status to include the words "Michael Jackson" at a rate of 100,000 tweets per hour.^[67] The current record as of August 3, 2013, was set in Japan, with 143,199 tweets per second during a television screening of the movie Castle in the Sky^[68] (beating the previous record of 33,388, also set by Japan for the television screening of the same movie).^[69]

The first unassisted off-Earth Twitter message was posted from the International Space Station by NASA astronaut T. J. Creamer on January 22, 2010.^[70] By late November 2010, an average of a dozen updates per day were posted on the astronauts' communal account, @NASA_Astronauts. NASA has also hosted over 25 "tweetups", events that provide guests with VIP access to NASA facilities and speakers with the goal of leveraging participants' social networks to further the outreach goals of NASA.

Twitter acquired application developer Atebits on April 11, 2010. Atebits had developed the Apple Design Award-winning Twitter client Tweetie for the Mac and iPhone. The application became the official Twitter client for the iPhone, iPad and Mac.^[71]

2010–2014

Logo used from 2012 to 2023

From September through October 2010, the company began rolling out "New Twitter", an entirely revamped edition of twitter.com. Changes included the ability to see pictures and videos without leaving Twitter itself by clicking on individual tweets which contain links to images and clips from a variety of supported websites, including YouTube and Flickr, and a complete overhaul of the interface, which shifted links such as '@mentions' and 'Retweets' above the Twitter stream, while 'Messages' and 'Log Out' became accessible via a black bar at the very top of twitter.com. As of November 1, 2010, the company confirmed that the "New Twitter experience" had been rolled out to all users. In 2019, Twitter was announced to be the 10th most downloaded mobile app of the decade, from 2010 to 2019.^[72]

On April 5, 2011, Twitter tested a new homepage and phased out the "Old Twitter".^[73] However, a glitch came about after the page was launched, so the previous "retro" homepage was still in use until the issues were resolved; the new homepage was reintroduced on April 20.^[74][75] On December 8, 2011, Twitter overhauled its website once more to feature the "Fly" design, which the service says is easier for new users to follow and promotes advertising. In addition to the Home tab, the Connect and Discover tabs were introduced along with a redesigned profile and timeline of Tweets. The site's layout has been compared to that of Facebook.^[76][77] On February 21, 2012, it was announced that Twitter and Yandex agreed to a partnership. Yandex, a Russian search engine, finds value within the partnership due to Twitter's real-time news feeds. Twitter's director of business development explained that it is important to have Twitter content where Twitter users go.^[78] On March 21, 2012, Twitter celebrated its sixth birthday by announcing that it had 140 million users, a 40% rise from September 2011, who were sending 340 million tweets per day.^[79][80]

On June 5, 2012, a modified logo was unveiled through the company blog, removing the text to showcase the slightly redesigned bird as the sole symbol of Twitter.^[81][82] On December 18, 2012, Twitter announced it had surpassed 200 million monthly active users.

On January 28, 2013, Twitter acquired Crashlytics in order to build out its mobile developer products.^[83] On April 18, 2013, Twitter launched a music app called Twitter Music for the iPhone.^[84] On August 28, 2013, Twitter acquired Trendrr,^[85] followed by the acquisition of MoPub on September 9, 2013.^[86] As of September 2013, the company's data showed that 200 million users sent over 400 million tweets daily, with nearly 60% of tweets sent from mobile devices.^[87]

During Super Bowl XLVII on February 3, 2013, when the power went out in the Mercedes-Benz Superdome Mondelez International, Kraft Foods vice president Lisa Mann was asked to tweet, "You can still dunk in the dark", referring to Oreo cookies. She approved, and as she told Ad Age in 2020, "literally the world [had] changed when I woke up the next morning." This became a milestone in the development of commenting daily on culture.^[88]

2014–2020

In April 2014, Twitter underwent a redesign that made the site resemble Facebook somewhat, with a profile picture and biography in a column left to the timeline, and a full-width header image with parallax scrolling effect.^{[lower-alpha 3]} That layout was used as the main for the desktop front end until July 2019, undergoing changes over time such as the removal of shortcut buttons to jump to the previous or next tweet in early 2017, and rounded profile pictures since June 2017.^[89][90][91]

In April 2015, the Twitter.com desktop homepage changed.^[92] Later in the year it became apparent that growth had slowed, according to Fortune,^[93] Business Insider,^[94] Marketing Land^[95] and other news websites including Quartz (in 2016).^[96]

On April 29, 2018, the first commercial tweet from space was sent by a private company Solstar utilizing solely commercial infrastructure during the New Shepard flight.^[97]

Since May 2018, tweet replies deemed by an algorithm to be detractive from the conversation are initially hidden and only loaded by actuating a "Show more replies" element at the bottom.^[98]

In 2019, Twitter released another redesign of its user interface.^[99]

2020–2021

The two tweets on May 26, 2020, from President Trump that Twitter had marked "potentially misleading" (inserting the blue warning icon and "Get the facts..." language) that led to the executive order

Twitter experienced considerable growth during the COVID-19 pandemic in 2020.^[100] The platform also was increasingly used for misinformation related to the pandemic.^[101] Twitter started marking tweets which contained misleading information, and adding links to fact-checks.^[102] In May 2020, Twitter moderators marked two tweets from U.S. President Donald Trump as "potentially misleading" and linked to a fact-check.^[103] Trump responded by signing an executive order to weaken Section 230 of the Communications Decency Act, which limits social media sites' liability for content moderation decisions.^{[104][105][106]} Twitter later banned Trump, claiming that he violated "the glorification of violence policy".^[107] The ban drew criticism from conservatives and European leaders, who saw it as an interference on freedom of speech.^[108]

On June 5, 2021, the Nigerian government issued an indefinite ban on Twitter usage in the country, citing "misinformation and fake news spread through it have had real world violent consequences",^[109] after the platform removed tweets made by the Nigerian President Muhammadu Buhari.^[110] Nigeria's ban was criticized by Amnesty International.^[111]

In 2021, Twitter began the research phase of Bluesky, an open source decentralized social media protocol where users can choose which algorithmic curation they want.^[112][113] The same year, Twitter also released Twitter Spaces, a social audio feature;^[114][115] "super follows", a way to subscribe to creators for exclusive content;^[116] and a beta of "ticketed Spaces", which makes access to certain audio rooms paid.^[117] Twitter unveiled a redesign in August 2021, with adjusted colors and a new Chirp font, which improves the left-alignment of most Western languages.^[118]

Since 2022

In June 2022, Twitter announced a partnership with e-commerce giant Shopify, and its plans to launch a sales channel app for U.S. Shopify merchants.^[119]

On August 23, 2022, the contents of a whistleblower complaint by former information security head Peiter Zatko to the United States Congress were published. Zatko had been fired by Twitter in January 2022. The complaint alleges that Twitter failed to disclose several data breaches, had negligent security measures, violated United States securities regulations, and broke the terms of a previous settlement with the Federal Trade Commission over the safeguarding of user data. The report also claims that the Indian government forced Twitter to hire one of its agents to gain direct access to user data.^[120]

Post-acquisition

Main article: Twitter under Elon Musk

A week after the takeover, Musk revamped Twitter Blue, increasing its price to $8 per month and adding new features, including the "blue checkmark" verification that had previously been reserved for high-profile confirmed users. This plan was criticized by several outlets, fearing that the potential for misinformation would increase since anyone could pay to appear to be verified through Twitter Blue. Musk opted to delay the changes to blue checkmarks until after the 2022 United States elections over these concerns and stated that accounts that were faking identities, outside of parody accounts, would be terminated.^[121] Following the takeover, various brands and companies paused advertising on the platform.^[122]

In December 2022, Musk provided internal documentation to a number of independent journalists and writers including Matt Taibbi, Bari Weiss and Michael Shellenberger which was then publicly disseminated through a series of posts known as the "Twitter Files". The files describe internal discussion as related to Twitter's moderation steps in events such as the breaking of the Hunter Biden laptop controversy, shadow banning of some conservative commentators' accounts, and the decision to block Trump. While some on the right saw the documents as evidence of Twitter's liberal bias and hostility to free speech, many people on the left described them as a reflection of how difficult it is for social media platforms to make tough decisions about content moderation.^{[123][124][125]}

The Twitter accounts of a number of journalists were permanently suspended on December 15, 2022. These journalists, including Mashable's Matt Binder, Aaron Rupar, Ryan Mac of The New York Times, and CNN's Donie O'Sullivan, had covered Twitter and recently had written articles about Musk's takeover. Several of the reporters had tweeted about the controversy over Musk's banning of ElonJet and other accounts which track private jets.^[126][127] In response, Twitter Head of Trust and Safety Ella Irwin told The Verge that "we will suspend any accounts that violate our privacy policies and put other users at risk".^[128] Binder denied violating any of Twitter's policies, saying: "I did not share any location data, as per Twitter's new terms. Nor did I share any links to ElonJet or other location tracking accounts."^[129] European Union (EU) commissioner Vera Jourova stated that the EU Digital Services Act insists on media freedom. Jourova stated: "Elon Musk should be aware of that. There are red lines. And sanctions, soon."^[130] On December 17, 2022, several banned journalists' accounts were reinstated.^[131]

On December 18, 2022, Twitter announced that it was banning users linking to what was referred to as "prohibited platforms" including Facebook, Mastodon, Instagram and Truth Social, as well as third party link aggregators such as Linktree.^[132] Six hours after his announcement, Musk tweeted: "Going forward, there will be a vote for major policy changes. My apologies. Won't happen again." It is unclear if the earlier policy will be reversed. He also tweeted a poll asking users if he should remain as head of Twitter promising to abide by its results,^[133][134] in which 57.5% of the 17.5m votes were in favour of him stepping down.^[135] On December 20, 2022, Musk announced he would step down as CEO once a replacement had been found, and that he would then run the software and servers teams.^[29] Linda Yaccarino resigned from NBCUniversal on May 12, 2023, and on the same day Musk announced that Yaccarino would be the new chief executive officer of X Corp. and Twitter.^[136]

Since the acquisition, Twitter's stability has been called into question, with reports that Musk had "disconnected one of the more sensitive server racks" in a data center used by Twitter, just before a large-scale outage of the platform occurred in December 2022.^[137] In 2023, many alternative third-party Twitter clients including Tweetbot, stopped working without warning and without any apparent explanation from the company.^[138] Later, Twitter announced it was "enforcing long-standing API rules" and retroactively changed its Developer Terms of Service afterward to justify its unannounced banning of third-party Twitter clients.^[139]

In January 2023, The Guardian stated that Twitter reportedly needed to pay $300m in interest on debts, and a default could lead to bankruptcy. Advertising revenue, comprising over 90% of Twitter's turnover had fallen strongly. This happened because clients were concerned about moderation standards after Musk's takeover and also due to a miscalculated restart of its subscription service, leading to many falsely "verified" corporate accounts. Advertising had not improved and Twitter's daily revenue had fallen by 40%.^[140]

Linda Yaccarino, current CEO of Twitter

On March 23, 2023, Twitter announced that "legacy" verification badges would be removed starting on April 1: this date passed with no change before Twitter and Musk both announced (at different dates) that the removal date for checkmarks from non-paying accounts was April 20.^[141] The only way to acquire a blue checkmark would be through the paid Twitter Blue subscription.^[142]

As of mid-April, Twitter no longer allows searching tweets without an account.^[143][144] On June 30, 2023, Twitter blocked unregistered users from viewing tweets or profiles on the platform. This not only blocked unregistered users from accessing tweets, but it also prevented search engines from indexing its pages, resulting in a loss of over half of its indexed URLs in the past few days.^[145] Musk described the change as a "temporary emergency measure" on his own profile, blaming "several hundred organizations" for scraping data from the site.^[146][147] The following day further measures were implemented with temporary limits to the amount of tweets a user can see per day, with verified users having 6,000, unverified users having 600 and newly created users having 300.^[148] This limit was later increased to 8,000 for verified users, 800 for unverified users, and 400 for new accounts.^[149] Around three hours after this announcement, the limit was once again raised to 10,000 for verified users, 1,000 for unverified users, and 500 for new accounts.^[149] After July 4, Twitter removed the block for unregistered users to view tweets and profiles; however, the ability to view threads and more than one tweet in a single page still requires an account.^[150]

Analysis conducted by research firm Sensor Tower found that global active daily users of the platform via mobile apps had steadily declined during the year after Musk acquired the company, down 16% by September 2023, while the metric showed positive growth for five other major social media platforms.^[151] With a surge in the number of users trying to migrate from Twitter to other platforms, people have raised questions regarding what migration patterns are, how different platforms impact user behaviors, and how migrated users settle in during the migration process. While Mastodon, Bluesky, and Threads are considered the main alternatives for this migration phenomenon, the enduring appeal of established platforms like Twitter poses a significant challenge in user retention for these new alternative platforms.^[152][153] By December 2023, loss of advertising revenue had become so extreme that public media speculation suggested that Twitter would become bankrupt. Accusations of uncontrolled antisemitism on the platform led to many advertisers withdrawing.^[154][155]

In January 2024, X announced plans to introduce a peer-to-peer payments system later. Musk said he anticipated a mid-year introduction. He reaffirmed comments he and investor Cathie Wood made late last year that the company had licenses to operate in a majority of states, although available public data indicated the number was 15 at the time of the announcement, with two major states, California and New York, not included. The move came after Musk's previous statement that X should become an app that would allow users to "conduct [their] entire financial world". Earlier in 2023, the company partnered with brokerage eToro to offer trading of different financial products.^{[156][157][158]}

Rebrand to X

Initial logo of X

Following Twitter's change in ownership, speculations of an upcoming rebrand began as Musk began referring to the platform as "X/Twitter"^{[159][160][161]} and "X (Twitter)",^[162] and renamed several features, including Birdwatch to Community Notes^[163] and Quote Tweets to Quotes.^[164] On July 23, 2023, Musk teased that the platform would receive a rebrand,^[165] beginning with a new logo based around the letter X. He later tweeted that if he found a logo, the rebrand would be implemented the next day. Subsequently, he tweeted a video of the Twitter logo glitching into an abstract X logo.^[166]

Later that day, Musk confirmed the rebrand, which started when the x.com domain (formerly associated with PayPal) began redirecting to Twitter;^[167] the logo was changed from the bird to the X the next day,^[168] and the platform's official main and associated accounts also began using the letter X within their handles.^[169] The @x handle was originally owned by photographer Gene X Hwang, who registered it in 2007. Hwang had expressed willingness to sell the handle, but received an email on July 25, 2023, stating that the company was taking it. He was offered some X merchandise and a meeting with the company's leaders, but no financial benefits.^[170] The Android app's name and icon were changed to X on Google Play by July 27; the same change went live on the App Store on July 31 after Apple granted an exception to its minimum character length of 2.^{[171][172][173]} Around that time, some more elements of the Twitter branding were removed from the web version, including tweets being renamed to "posts".^[174]

The rebrand was described as unusual, given that Twitter's brand was already strong internationally, with words like "tweet" having entered common language.^[175] The rebranding has been criticized on the basis that the trademarkability of the name and logo is weak: there are almost 900 companies in the U.S. that own an X trademark,^[176] including an existing social media-related logo owned by Meta Platforms.^[177] The X logo uses a blackboard bold X, a character that has appeared in mathematical textbooks since the 1970s and that is included in Unicode as U+1D54F 𝕏 MATHEMATICAL DOUBLE-STRUCK CAPITAL X;^[178] one user of the platform also noted a similarity of the logo to the lowercase x in the Monotype font Special Alphabets 4, though Monotype has confirmed that they are not exactly alike.^[179]

A few days after the rebrand took effect, an AP Stylebook update recommended that journalists refer to the platform as "X, formerly known as Twitter".^[4] In September 2023, Ad Age, citing The Harris Poll, noted that the rebranding had not publicly caught on, with the majority of users as well as notable brands still referring to X as "Twitter".^[180]

Posts

The X account page for Wikipedia as of November 2023

Posts (or tweets) are publicly visible by default, but senders can restrict message delivery to only their followers. Users can mute users they do not wish to interact with, block accounts from viewing their posts, and remove accounts from their followers list.^{[181][182][183]} Users can post via the Twitter website, compatible external applications (such as for smartphones), or by Short Message Service (SMS) available in certain countries.^[184] Users may subscribe to other users' posts—this is known as "following" and subscribers are known as "followers"^[185] or "tweeps", a portmanteau of Twitter and peeps.^[186] Individual posts can be forwarded by other users to their own feed, a process known as a "repost" or "retweet". In 2015, Twitter launched "quote tweet" (originally called "retweet with comment"),^[187] a feature that allows users to add a comment to their post, imbedding one post in the other.^[188] Users can also "like" (formerly "favorite") individual tweets.^[189]

The counters for "likes", "retweets/reposts", and replies appear next to the respective buttons in timelines such as on profile pages and search results. Counters for likes and reposts exist on a post's standalone page too. Since September 2020, quote tweets, formerly known as "retweet with comment", have their own counter on their post page.^[187] Until the legacy desktop front end that was discontinued in 2020, a row with miniature profile pictures of up to ten liking or retweeting users was displayed (earliest documented implementation in December 2011 overhaul), as well as a tweet reply counter next to the according button on a tweet's page.^[190][191]

Twitter allows users to update their profile via their mobile phones either by text messaging or by apps released for certain smartphones and tablets.^[192] Twitter has been compared to a web-based Internet Relay Chat (IRC) client.^[193] Twitter announced in a tweet on September 1, 2022, that the ability to edit a tweet was being tested for select users. The company said the feature was being tested first to determine whether it could be abused. Editing would be allowed for 30 minutes, and previous versions of an edited post would be available. Eventually, all Twitter Blue subscribers would be able to use the feature.^[194]

Users can group posts together by topic or type by use of hashtags – words or phrases prefixed with a "#" sign. Similarly, the "@" sign followed by a username is used for mentioning or replying to other users.^[195] In 2014, Twitter introduced hashflags, special hashtags that automatically generate a custom emoji next to them for a certain period of time.^[196] Hashflags may be generated by Twitter themselves^[197] or be purchased by corporations.^[198] To repost a message from another X user and share it with one's own followers, a user can click the repost button within the post. Users can reply to other accounts' replies. Users can hide replies to their messages and select who can reply to each of their tweets before sending them: anyone, accounts who follow the poster, specific accounts, and none.^[199][200]

The original, strict 140 character limit was gradually relaxed. In 2016, Twitter announced that attachments, links, and media such as photos, videos, and the person's handle, would no longer count; a user photo post used to count for around 24 characters.^[201][202] In 2017, Twitter handles were similarly excluded.^[203] The same year, Twitter doubled its historical 140-character-limitation to 280.^[204] Under the new limit, glyphs are counted as a variable number of characters, depending upon the script they are from.^[204] In 2023, Twitter announced that Twitter Blue users could create posts with up to 4,000 characters in length.^[205]

t.co is a URL shortening service created by Twitter.^[206] It is only available for links posted to Twitter and not available for general use.^[206] All links posted to Twitter use a t.co wrapper.^[207] Twitter intended the service to protect users from malicious sites,^[206] and to use it to track clicks on links within tweets.^[206][208] Twitter had previously used the services of third parties TinyURL and bit.ly.^[209]

In June 2011, Twitter announced its own integrated photo-sharing service that enables users to upload a photo and attach it to a Tweet right from Twitter.com.^[210] Users now also have the ability to add pictures to Twitter's search by adding hashtags to the tweet.^[211] Twitter also plans to provide photo galleries designed to gather and syndicate all photos that a user has uploaded on Twitter and third-party services such as TwitPic.^[211] On March 29, 2016, Twitter introduced the ability to add a caption of up to 480 characters to each image attached to a tweet,^[212][213] accessible via screen reading software or by hovering the mouse above a picture inside TweetDeck. In April 2022, Twitter made the ability to add and view captions globally available. Descriptions can be added to any uploaded image with a limit of 1000 characters. Images that have a description will feature a badge that says ALT in the bottom left corner, which will bring up the description when clicked.^[214]

In 2015, Twitter began to roll out the ability to attach poll questions to tweets. Polls are open for up to 7 days, and voters are not personally identified.^[215] In Twitter's early years, users could communicate with Twitter using SMS. Twitter discontinued this feature in most countries in April 2023, after hackers had exposed vulnerabilities in the feature.^[216][217]

Non-tweet content

In 2016, Twitter began to place a larger focus on live streaming video programming, hosting various events including streams of the Republican and Democratic conventions during the U.S. presidential campaign,^[218] and winning a bid for non-exclusive streaming rights to ten NFL games in 2016.^[219][220] During an event in New York in May 2017, Twitter announced that it planned to construct a 24-hour streaming video channel hosted within the service, featuring content from various partners.^[219][221] Twitter announced a number of new and expanded partnerships for its streaming video services at the event, including Bloomberg, BuzzFeed, Cheddar, IMG Fashion, Live Nation Entertainment, Major League Baseball, MTV and BET, NFL Network, the PGA Tour, The Players' Tribune, Ben Silverman and Howard T. Owens' Propagate, The Verge, Stadium and the WNBA.^[222] as of the first quarter of 2017, Twitter had over 200 content partners, who streamed over 800 hours of video over 450 events.^[222]

Twitter Spaces is a social audio feature that enables users to host or participate in a live-audio virtual environment called space for conversation. A maximum of 13 people are allowed onstage. The feature was initially limited to users with at least 600 followers, but since October 2021, any Twitter user can create a Space.^[223]

In March 2020, Twitter began to test a stories feature known as "fleets" in some markets,^[224][225] which officially launched on November 17, 2020.^[226][227] Fleets could contain text and media, are only accessible for 24 hours after they are posted, and are accessed within the Twitter app;^[224] Twitter announced it would start implementing advertising into fleets in June 2021.^[228] Fleets were removed in August 2021; Twitter had intended for fleets to encourage more users to tweet regularly, but instead they were generally used by already-active users.^[229]

Curation

Trending topics

Main article: Twitter trends

Countries and cities with local trending topics in Twitter

A word, phrase, or topic that is mentioned at a greater rate than others is said to be a "trending topic". Trending topics become popular either through a concerted effort by users or because of an event that prompts people to talk about a specific topic.^[230] These topics help Twitter and their users to understand what is happening in the world and what people's opinions are about it.^[231] The Twitter web interface displays a list of trending topics on a sidebar on the home page, along with sponsored content.

Trending topics are sometimes the result of concerted efforts and manipulations by fans of certain celebrities or cultural phenomena, particularly musicians like Lady Gaga, Justin Bieber, Rihanna and One Direction, and novel series Twilight and Harry Potter. Twitter has altered the trend algorithm in the past to prevent manipulation of this type with limited success.^[232] Twitter also censors trending hashtags that are claimed to be abusive or offensive. Twitter censored the #Thatsafrican^[233] and #thingsdarkiessay hashtags after users complained that they found the hashtags offensive.^[234]

Algorithm

On October 21, 2021, report based on a "long-running, massive-scale randomized experiment" that analyzed "millions of tweets sent between 1 April and 15 August 2020", found that Twitter's machine learning recommendation algorithm amplified right-leaning politics on personalized user Home timelines.^{[241]: 1 [242]} The report compared seven countries with active Twitter users where data was available—Germany, Canada, the United Kingdom, Japan, France, and Spain—and examined Tweets "from major political groups and politicians".^[241]: 4 Researchers used the 2019 Chapel Hill Expert Survey (CHESDATA) to position parties on political ideology within each country.^[241]: 4 The "machine learning algorithms"—introduced by Twitter in 2016—personalized 99% of users' feeds by displaying Tweets—even older Tweets and Retweets from accounts the user had not directly followed—but that the algorithm had "deemed relevant" to the users' past preferences.^[241]: 4 Twitter randomly chose 1% of users whose Home timelines displayed content in reverse-chronological order from users they directly followed.^[241]: 2

Mobile

Twitter has mobile apps for iPhone, iPad, and Android.^[243] In April 2017, Twitter introduced Twitter Lite, a progressive web app designed for regions with unreliable and slow Internet connections, with a size of less than one megabyte, designed for devices with limited storage capacity.^[244][245]

Premium (formerly Twitter Blue)

On June 3, 2021, Twitter announced a paid subscription service called Twitter Blue. Following Twitter's ongoing "X" name change, the subscription was rebranded to X Blue (or simply Blue), and, on August 5, 2023, was rebranded to X Premium (or simply Premium).^[246][247] The subscription provides additional premium features to the service.^[248][249] In November 2023 a "Premium+" subscription was launched, with a higher monthly fee giving benefits such as the omission of adverts.^[250]

User monetization

In June 2021, the company opened applications for its premium subscription options called Super Follows. This lets eligible accounts charge $2.99, $4.99 or $9.99 per month to subscribe to the account.^[266] The launch only generated about $6,000 in its first two weeks.^[267] In 2023, the Super Follows feature was rebranded as simply "subscriptions", allowing users to publish exclusive long-form posts and videos for their subscribers; the pivot in marketing was reportedly intended to help compete with Substack.^[268]

In May 2021, Twitter began testing a Tip Jar feature on its iOS and Android clients. The feature allows users to send monetary tips to certain accounts, providing a financial incentive for content creators on the platform. The Tip Jar is optional and users can choose whether or not to enable tips for their account.^[269] On September 23, 2021, Twitter announced that it will allow users to tip users on the social network with bitcoin. The feature will be available for iOS users. Previously, users could tip with fiat currency using services such as Square's Cash App and PayPal's Venmo. Twitter will integrate the Strike bitcoin lightning wallet service. It was noted that at this current time, Twitter will not take a cut of any money sent through the tips feature.^[270]

On August 27, 2021, Twitter rolled out Ticketed Spaces, which let Twitter Spaces hosts charge between $1 and $999 for access to their rooms.^[271] In April 2022, Twitter announced that it will partner with Stripe, Inc. for piloting cryptocurrency payouts for limited users in the platform. Eligible users of Ticketed Spaces and Super Follows will be able to receive their earnings in the form of USD coin, a stablecoin whose value is that of the U.S. dollar. Users can also hold their earnings in crypto wallets, and then exchange them into other cryptocurrencies.^[272]

E-commerce

From 2014 to 2017, Twitter offered a "Buy button" feature, allowing tweets to embed products that could be purchased from within the service. Users could also add their billing and shipping information directly to their accounts. The buy button's platform partners at launch included Stripe, Gumroad, Musictoday, and The Fancy,^[273]

In July 2021, Twitter began testing a "Shop module" for iOS users in the United States, allowing accounts associated with brands to display a carousel of cards on their profiles showcasing products. Unlike the Buy button, where order fulfillment was handed from within Twitter, these cards are external links to online storefronts from which the products may be purchased.^[274] In March 2022, Twitter expanded the test to allow companies to showcase up to 50 products on their profiles.^[275]

In November 2021, Twitter introduced support for "shoppable" live streams, in which brands can hold streaming events that similarly display banners and pages highlighting products that are featured in the presentation.^[276]

Demographics

In 2009, Twitter was mainly used by older adults who might not have used other social sites before Twitter.^[279] According to comScore only 11% of Twitter's users were aged 12 to 17.^[279] According to a study by Sysomos in June 2009, women made up a slightly larger Twitter demographic than men—53% over 47%. It also stated that 5% of users accounted for 75% of all activity.^[280] According to Quancast, 27 million people in the US used Twitter in September 2009; 63% of Twitter users were under 35 years old; 60% of Twitter users were Caucasian, but a higher than average (compared to other Internet properties) were African American/black (16%) and Hispanic (11%); 58% of Twitter users have a total household income of at least US$60,000.^[281] The prevalence of African American Twitter usage and in many popular hashtags has been the subject of research studies.^[282][283]

Twitter grew from 100 million monthly active users (MAUs) in September 2011,^[284] to 255 million in March 2014,^[285] and more than 330 million in early 2019.^{[286][287][8]} In 2013, there were over 100 million users actively using Twitter daily and about 500 million tweets every day.^[288] A 2016 Pew research poll found that Twitter is used by 24% of all online US adults. It was equally popular with men and women (24% and 25% of online Americans respectively), but more popular with younger generations (36% of 18–29 year olds).^[289] A 2019 survey conducted by the Pew Foundation found that Twitter users are three times as likely to be younger than 50 years old, with the median age of adult U.S. users being 40. The survey found that 10% of users who are most active on Twitter are responsible for 80% of all tweets.^[290]

Content

Content of tweets according to Pear Analytics in August 2009

  News (3.6%)

  Spam (3.8%)

  Self-promotion (5.9%)

  Pointless babble (40.1%)

  Conversational (37.6%)

  Pass-along value (8.7%)

San Antonio-based market-research firm Pear Analytics analyzed 2,000 tweets (originating from the United States and in English) over a two-week period in August 2009 from 11:00 am to 5:00 pm (CST) and separated them into six categories.^[291] Pointless babble made up 40%, with 38% being conversational. Pass-along value had 9%, self-promotion 6% with spam and news each making 4%.

Despite Jack Dorsey's own open contention that a message on Twitter is "a short burst of inconsequential information", social networking researcher danah boyd responded to the Pear Analytics survey by arguing that what the Pear researchers labeled "pointless babble" is better characterized as "social grooming" or "peripheral awareness" (which she justifies as persons "want[ing] to know what the people around them are thinking and doing and feeling, even when co-presence isn't viable").^[292] Similarly, a survey of Twitter users found that a more specific social role of passing along messages that include a hyperlink is an expectation of reciprocal linking by followers.^[293]

Levels of use

According to research published in April 2014, around 44% of user accounts have never tweeted.^[294] About 22% of Americans say they have ever used Twitter, according to a 2019 Pew Research Center survey.^[295] In 2009, Nielsen Online reported that Twitter had a user-retention rate of forty percent. Many people stop using the service after a month; therefore the site may potentially reach only about ten percent of all Internet users.^[296] Noting how demographics of Twitter users differ from the average Americans, commentators have cautioned against media narratives that treat Twitter as representative of the population,^[297] adding that only 10% of users Tweet actively, and that 90% of Twitter users have Tweeted no more than twice. In 2016, shareholders sued Twitter, alleging it "artificially inflated its stock price by misleading them about user engagement". The company announced on September 20, 2021, that it would pay $809.5 million to settle this class-action lawsuit.^[298]

User engagement

User engagement is usually measured by the number of likes, replies and reposts. A 2023 study showed that retweets are more likely to contain positive content and address larger audiences using the first-person pronoun "we". Replies, on the other hand, are more likely to contain negative content and address individuals using the second-person pronoun "you" and the third-person pronouns "he" or "she". While influencers with many followers tend to post positive messages, often using the word "love" when addressing larger audiences, users with less followers tend to engage in interpersonal conversations to provoke user engagement.^[299]

Revenue sources

Twitter's San Francisco headquarters located at 1355 Market Street

On April 13, 2010, Twitter announced plans to offer paid advertising for companies that would be able to purchase "promoted tweets" to appear in selective search results on the Twitter website, similar to Google Adwords' advertising model.^[310][311] Users' photos can generate royalty-free revenue for Twitter, and an agreement with World Entertainment News Network (WENN) was announced in May 2011.^[312] Twitter generated an estimated US$139.5 million in advertising sales during 2011.^[313]

In June 2011, Twitter announced that it would offer small businesses a self-service advertising system.^[314] The self-service advertising platform was launched in March 2012 to American Express card members and merchants in the U.S. on an invite-only basis.^[315] To continue their advertising campaign, Twitter announced on March 20, 2012, that promoted tweets would be introduced to mobile devices.^[316] In April 2013, Twitter announced that its Twitter Ads self-service platform, consisting of promoted tweets and promoted accounts, was available to all U.S. users without an invite.^[315]

On August 3, 2016, Twitter launched Instant Unlock Card, a new feature that encourages people to tweet about a brand in order to earn rewards and utilize the social media network's conversational ads. The format itself consists of images or videos with call-to-action buttons and a customizable hashtag.^[317]

Advertising bans

In October 2017, Twitter banned the Russian media outlets RT and Sputnik from advertising on their website following the conclusions of the U.S. national intelligence report the previous January that both Sputnik and RT had been used as vehicles for Russia's interference in the 2016 US presidential election.^[318] Maria Zakharova for the Russian foreign ministry said the ban was a "gross violation" by the US of free speech.^[319]

In October 2019, Twitter announced it would stop running political ads on its ad platform effective November 22. This resulted from several spurious claims made by political ads. Company CEO Dorsey clarified that internet advertising had great power and was extremely effective for commercial advertisers, the power brings significant risks to politics where crucial decisions impact millions of lives.^[320] The company reversed the ban in August 2023,^[321] publishing criteria governing political advertising which do not allow the promotion of false or misleading content, and requiring advertisers to comply with laws, with compliance being the sole responsibility of the advertiser.^[322]

In April 2022, Twitter announced a ban on "misleading" advertisements that go against "the scientific consensus on climate change". While the company did not give full guidelines, it stated that the decisions would be made with the help of "authoritative sources", including the Intergovernmental Panel on Climate Change.^[323]

Fines

Twitter, and X, has been fined several times for non-compliance with laws and regulations. On May 25, 2022, Twitter was fined $150 million by the Federal Trade Commission and the United States Department of Justice for collecting users' contact details and using them for targeted advertising.^[324][325]

In August 2023, Twitter was fined US$350,000 for failing to meet two deadlines to comply with a U.S. Department of Justice search warrant for the account of former president Donald Trump.^[326][327]

In October 2023, X was fined AU$610,500 by Australia's e-Safety Commission for failing to properly disclose information about how it polices child abuse content.^[328][329]

As of 2023 other fines were being considered.^[330][331]

Implementation

Twitter relies on open-source software.^[332] The Twitter Web interface uses the Ruby on Rails framework,^[333] deployed on a performance enhanced Ruby Enterprise Edition implementation of Ruby.^[334]

In the early days of Twitter, tweets were stored in MySQL databases that were temporally sharded (large databases were split based on time of posting). After the huge volume of tweets coming in caused problems reading from and writing to these databases, the company decided that the system needed re-engineering.^[68]

From Spring 2007 to 2008, the messages were handled by a Ruby persistent queue server called Starling.^[335] Since 2009, implementation has been gradually replaced with software written in Scala.^[336] The switch from Ruby to Scala and the JVM has given Twitter a performance boost from 200 to 300 requests per second per host to around 10,000–20,000 requests per second per host. This boost was greater than the 10x improvement that Twitter's engineers envisioned when starting the switch. The continued development of Twitter has also involved a switch from monolithic development of a single app to an architecture where different services are built independently and joined through remote procedure calls.^[68]

As of April 6, 2011, Twitter engineers confirmed that they had switched away from their Ruby on Rails search stack to a Java server they call Blender.^[337]

Individual tweets are registered under unique IDs called snowflakes, and geolocation data is added using 'Rockdove'. The URL shortener t.co then checks for a spam link and shortens the URL. Next, the tweets are stored in a MySQL database using Gizzard, and the user receives an acknowledgement that the tweets were sent. Tweets are then sent to search engines via the Firehose API. The process is managed by FlockDB and takes an average of 350 ms.^[332]

On August 16, 2013, Raffi Krikorian, Twitter's vice president of platform engineering, shared in a blog post that the company's infrastructure handled almost 143,000 tweets per second during that week, setting a new record. Krikorian explained that Twitter achieved this record by blending its homegrown and open source technologies.^[68][338]

API and developer platform

Twitter was recognized for having one of the most open and powerful developer APIs of any major technology company.^[339] The service's API allows other web services and applications to integrate with Twitter.^[340] Developer interest in Twitter began immediately following its launch, prompting the company to release the first version of its public API in September 2006.^[341] The API quickly became iconic as a reference implementation for public REST APIs and is widely cited in programming tutorials.^[342]

From 2006 until 2010, Twitter's developer platform experienced strong growth and a highly favorable reputation. Developers built upon the public API to create the first Twitter mobile phone clients as well as the first URL shortener. Between 2010 and 2012, however, Twitter made a number of decisions that were received unfavorably by the developer community.^[343] In 2010, Twitter mandated that all developers adopt OAuth authentication with just 9 weeks of notice.^[344] Later that year, Twitter launched its own URL shortener, in direct competition with some of its most well-known third-party developers.^[345] And in 2012, Twitter introduced stricter usage limits for its API, "completely crippling" some developers.^[346][347] While these moves successfully increased the stability and security of the service, they were broadly perceived as hostile to developers, causing them to lose trust in the platform.^[348]

In July 2020, Twitter released version 2.0 of the public API^[349] and began showcasing Twitter apps made by third-party developers on its Twitter Toolbox section in April 2022.^[350]

In January 2023, Twitter ended third party access to its APIs, forcing all third party Twitter clients to shut down.^[351] This was controversial among the developer community, as many third-party apps predated the company's official apps, and the change was not announced beforehand. Twitterrific's Sean Heber confirmed in a blog post that the 16-year-old app has been discontinued. "We are sorry to say that the app's sudden and undignified demise is due to an unannounced and undocumented policy change by an increasingly capricious Twitter – a Twitter that we no longer recognize as trustworthy nor want to work with any longer."^[352]

In February 2023, Twitter announced it would be ending free access to Twitter API, and began offering paid tier plans with a more limited access.^[353]

Innovators patent agreement

On April 17, 2012, Twitter announced it would implement an "Innovators Patent Agreement" which would obligate Twitter to only use its patents for defensive purposes.^[354]

Open source

Twitter has a history of both using and releasing open-source software while overcoming technical challenges of their service.^[355] A page in their developer documentation thanks dozens of open-source projects which they have used, from revision control software like Git to programming languages such as Ruby and Scala.^[356] Software released as open source by the company includes the Gizzard Scala framework for creating distributed datastores, the distributed graph database FlockDB, the Finagle library for building asynchronous RPC servers and clients, the TwUI user interface framework for iOS, and the Bower client-side package manager.^[357] The popular Bootstrap frontend framework was also started at Twitter and is 10th most popular repository on GitHub.^[358]

On March 31, 2023, Twitter released the source code for Twitter's recommendation algorithm,^[359] which determines what tweets show up on the user's personal timeline, to GitHub. According to Twitter's blog post: "We believe that we have a responsibility, as the town square of the internet, to make our platform transparent. So today we are taking the first step in a new era of transparency and opening much of our source code to the global community."^[360] CEO Elon Musk had been promising the move for a while — on March 24, 2022, before he owned the site, he polled his followers about whether Twitter's algorithm should be open source, and around 83 percent of the responses said "yes". In February, he promised it would happen within a week before pushing back the deadline to March 31 earlier this month.^[361]

Interface

Twitter introduced the first major redesign of its user interface in September 2010, adopting a dual-pane layout with a navigation bar along the top of the screen, and an increased focus on the inline embedding of multimedia content. Critics considered the redesign an attempt to emulate features and experiences found in mobile apps and third-party Twitter clients.^{[362][363][364][365]}

The new layout was revised in 2011 with a focus on continuity with the web and mobile versions, introducing "Connect" (interactions with other users such as replies) and "Discover" (further information regarding trending topics and news headlines) tabs, an updated profile design, and moving all content to the right pane (leaving the left pane dedicated to functions and the trending topics list).^[366] In March 2012, Twitter became available in Arabic, Farsi, Hebrew and Urdu, the first right-to-left language versions of the site.^[367] In 2023 the Twitter Web site listed 34 languages supported by Twitter.com.^[368]

In September 2012, a new layout for profiles was introduced, with larger "covers" that could be customized with a custom header image, and a display of the user's recent photos posted.^[369] The "Discover" tab was discontinued in April 2015,^[370] and was succeeded on the mobile app by an "Explore" tab—which features trending topics and moments.^[371]

In September 2018, Twitter began to migrate selected web users to its progressive web app (based on its Twitter Lite experience for mobile web), reducing the interface to two columns. Migrations to this iteration of Twitter increased in April 2019, with some users receiving it with a modified layout.^[372][373]

In July 2019, Twitter officially released this redesign, with no further option to opt-out while logged in. It is designed to further-unify Twitter's user experience between the web and mobile application versions, adopting a three-column layout with a sidebar containing links to common areas (including "Explore" that has been merged with the search page) which previously appeared in a horizontal top bar, profile elements such as picture and header images and biography texts merged into the same column as the timeline, and features from the mobile version (such as multi-account support, and an opt-out for the "top tweets" mode on the timeline).^[374][375]

Security

In response to early Twitter security breaches, the United States Federal Trade Commission (FTC) brought charges against the service; the charges were settled on June 24, 2010. This was the first time the FTC had taken action against a social network for security lapses. The settlement requires Twitter to take a number of steps to secure users' private information, including maintenance of a "comprehensive information security program" to be independently audited biannually.^[376]

After a number of high-profile hacks of official accounts, including those of the Associated Press and The Guardian,^[377] in April 2013, Twitter announced a two-factor login verification as an added measure against hacking.^[378]

On July 15, 2020, a major hack of Twitter affected 130 high-profile accounts, both verified and unverified ones such as Barack Obama, Bill Gates, and Elon Musk; the hack allowed bitcoin scammers to send tweets via the compromised accounts that asked the followers to send bitcoin to a given public address, with the promise to double their money.^[379] Within a few hours, Twitter disabled tweeting and reset passwords from all verified accounts.^[379] Analysis of the event revealed that the scammers had used social engineering to obtain credentials from Twitter employees to access an administration tool used by Twitter to view and change these accounts' personal details as to gain access as part of a "smash and grab" attempt to make money quickly, with an estimated US$120,000 in bitcoin deposited in various accounts before Twitter intervened.^[380] Several law enforcement entities including the FBI launched investigations into the attack.^[381]

On August 5, 2022, Twitter disclosed that a bug introduced in a June 2021 update to the service allowed threat actors to link email addresses and phone numbers to twitter user's accounts.^[382][383] The bug was reported through Twitter's bug bounty program in January 2022 and subsequently fixed. While Twitter originally believed no one had taken advantage of the vulnerability, it was later revealed that a user on the online hacking forum Breached Forums had used the vulnerability to compile a list of over 5.4 million user profiles, which they offered to sell for $30,000.^[384][385] The information compiled by the hacker includes user's screen names, location and email addresses which could be utilised in phishing attacks or used to deanonymize accounts running under pseudonyms.

Outages

The Twitter fail whale error message

During an outage, Twitter users were at one time shown the "fail whale" error message image created by Yiying Lu,^[386] illustrating eight orange birds using a net to hoist a whale from the ocean captioned "Too many tweets! Please wait a moment and try again."^[387] Web designer and Twitter user Jen Simmons was the first to coin the term "fail whale" in a September 2007 tweet.^[388][389] In a November 2013 Wired interview Chris Fry, VP of Engineering at that time, noted that the company had taken the "fail whale" out of production as the platform was now more stable.^[390] Twitter had approximately ninety-eight percent uptime in 2007 (or about six full days of downtime).^[391] The downtime was particularly noticeable during events popular with the technology industry such as the 2008 Macworld Conference & Expo keynote address.^[392][393]

Verified accounts

In June 2009, after being criticized by Kanye West and sued by Tony La Russa over unauthorized accounts run by impersonators, the company launched their "Verified Accounts" program.^[394][395] Twitter stated that an account with a "blue tick" verification badge indicates "we've been in contact with the person or entity the account is representing and verified that it is approved".^[396] In July 2016, Twitter announced a public application process to grant verified status to an account "if it is determined to be of public interest" and that verification "does not imply an endorsement".^{[397][398][399]} Verified status allows access to some features unavailable to other users, such as only seeing mentions from other verified accounts.^[400]

In November 2020, Twitter announced a relaunch of its verification system in 2021. According to the new policy, Twitter verifies six different types of accounts; for three of them (companies, brands, and influential individuals like activists), the existence of a Wikipedia page will be one criterion for showing that the account has "Off Twitter Notability".^[401] Twitter states that it will re-open public verification applications at some point in "early 2021".^[402]

In October 2022, after the takeover of Twitter by Elon Musk, it was reported that verification would instead be included in the paid Twitter Blue service, and that existing verified accounts would lose their status if they do not subscribe.^[403] On November 1, Musk confirmed that verification would be included in Blue in the future, dismissing the existing verification system as a "lords & peasants system".^{[251][252][253]} Following concerns over the possibility of impersonation, Twitter subsequently reimplemented a second "Official" marker, consisting of a grey tick and "Official" text displayed under the username, for high-profile accounts of "government and commercial entities".^[404][405]

In December 2022, the "Official" text was replaced by a gold checkmark for organizations, as well as a grey checkmark for government and multilateral accounts.^[406][407]

In March 2023, the gold checkmark was made available for organizations to purchase through the Verified Organizations program (formerly called Twitter Blue for Business).^[406][407]

Privacy

Tweets are public, but users can also send private "direct messages".^[408] Information about who has chosen to follow an account and who a user has chosen to follow is also public, though accounts can be changed to "protected" which limits this information (and all tweets) to approved followers.^[409] Twitter collects personally identifiable information about its users and shares it with third parties as specified in its privacy policy. The service also reserves the right to sell this information as an asset if the company changes hands.^[410] Advertisers can target users based on their history of tweets and may quote tweets in ads^[411] directed specifically to the user.

Twitter launched the beta version of their "Verified Accounts" service on June 11, 2009, allowing people with public profiles to announce their account name. The profile pages of these accounts display a badge indicating their status.^[412]

On December 14, 2010, the United States Department of Justice issued a subpoena directing Twitter to provide information for accounts registered to or associated with WikiLeaks.^[413] Twitter decided to notify its users and said in a statement, "... it's our policy to notify users about law enforcement and governmental requests for their information, unless we are prevented by law from doing so."^[408]

In May 2011, a claimant known as "CTB" in the case of CTB v Twitter Inc. took action against Twitter at the High Court of Justice of England and Wales,^[414] requesting that the company release details of account holders. This followed gossip posted on Twitter about professional footballer Ryan Giggs's private life. This led to the 2011 British privacy injunctions controversy and the "super-injunction".^[415] Tony Wang, the head of Twitter in Europe, said that people who do "bad things" on the site would need to defend themselves under the laws of their own jurisdiction in the event of controversy and that the site would hand over information about users to the authorities when it was legally required to do so.^[416] He also suggested that Twitter would accede to a UK court order to divulge names of users responsible for "illegal activity" on the site.^[417]

Twitter acquired Dasient, a startup that offers malware protection for businesses, in January 2012. Twitter announced plans to use Dasient to help remove hateful advertisers on the website.^[418] Twitter also offered a feature which would allow tweets to be removed selectively by country, before deleted tweets used to be removed in all countries.^[419][420] The first use of the policy was to block the account of German neo-Nazi group Besseres Hannover on October 18, 2012.^[421] The policy was used again the following day to remove anti-Semitic French tweets with the hashtag #unbonjuif ("a good Jew").^[422]

Followed the sharing of images showing the killing of American journalist James Foley in 2014, Twitter said that in certain cases it would delete pictures of people who had died after requests from family members and "authorized individuals".^[423][424]

In 2015, following updated terms of service and privacy policy, Twitter users outside the United States were legally served by the Ireland-based Twitter International Company instead of Twitter, Inc. The change made these users subject to Irish and European Union data protection laws^[425]

On April 8, 2020, Twitter announced that users outside of the European Economic Area or United Kingdom (thus subject to GDPR) will no longer be allowed to opt out of sharing "mobile app advertising measurements" to Twitter third-party partners.^[426]

On October 9, 2020, Twitter took additional steps to counter misleading campaigns ahead of the 2020 US Election. Twitter's new temporary update encouraged users to "add their own commentary" before retweeting a tweet, by making 'quoting tweet' a mandatory feature instead of optional. The social network giant aimed at generating context and encouraging the circulation of more thoughtful content.^[427] After limited results, the company ended this experiment in December 2020.^[428]

On May 25, 2022, Twitter was fined $150 million for collecting users' phone numbers and email addresses used for security and using them for targeted advertising, required to notify its users, and banned from profiting from 'deceptively collected data'.^[429]

Harassment

In August 2013, Twitter announced plans to introduce a "report abuse" button for all versions of the site following uproar, including a petition with 100,000 signatures, over Tweets that included rape and death threats to historian Mary Beard, feminist campaigner Caroline Criado-Perez and the member of parliament Stella Creasy.^{[430][431][432]} Twitter announced new reporting and blocking policies in December 2014,^{[433][434][435][436]} including a blocking mechanism devised by Randi Harper, a target of GamerGate.^{[437][438][439]} In February 2015, CEO Dick Costolo said he was 'frankly ashamed' at how poorly Twitter handled trolling and abuse, and admitted Twitter had lost users as a result.^[440]

As per a research study conducted by IT for Change on abuse and misogynistic trolling on Twitter directed at Indian women in public-political life, women perceived to be ideologically left-leaning, dissenters, Muslim women, political dissenters, and political commentators and women from opposition parties received a disproportionate amount of abusive and hateful messages on Twitter.^[441]

In 2016, Twitter announced the creation of the Twitter Trust & Safety Council to help "ensure that people feel safe expressing themselves on Twitter". The council's inaugural members included 50 organizations and individuals.^[442] The announcement of Twitter's "Trust & Safety Council" was met with objection from parts of its userbase.^[443][444] Critics accused the member organizations of being heavily skewed towards "the restriction of hate speech" and a Reason article expressed concern that "there's not a single uncompromising anti-censorship figure or group on the list".^[445][446]

Twitter banned 7,000 accounts and limited 150,000 more that had ties to QAnon on July 21, 2020. The bans and limits came after QAnon-related accounts began harassing other users through practices of swarming or brigading, coordinated attacks on these individuals through multiple accounts in the weeks prior. Those accounts limited by Twitter will not appear in searches nor be promoted in other Twitter functions. Twitter said they will continue to ban or limit accounts as necessary, with their support account stating "We will permanently suspend accounts Tweeting about these topics that we know are engaged in violations of our multi-account policy, coordinating abuse around individual victims, or are attempting to evade a previous suspension".^[447]

In September 2021, Twitter began beta testing a feature called Safety Mode.^[448] The functionality aims to limit unwelcome interactions through automated detection of negative engagements. If a user has Safety Mode enabled, authors of tweets that are identified by Twitter's technology as being harmful or exercising uninvited behavior will be temporarily unable to follow the account, send direct messages, or see tweets from the user with the enabled functionality during the temporary block period.^[449] Jarrod Doherty, senior product manager at Twitter, stated that the technology in place within Safety Mode assesses existing relationships to prevent blocking accounts that the user frequently interacts with.^[448]

Suspect and contested accounts

In January 2016, Twitter was sued by the widow of a U.S. man killed in the 2015 Amman shooting attack, claiming that allowing the Islamic State of Iraq and the Levant (ISIL) to continually use the platform, including direct messages in particular,^[450] constituted the provision of material support to a terrorist organization, which is illegal under U.S. federal law. Twitter disputed the claim, stating that "violent threats and the promotion of terrorism deserve no place on Twitter and, like other social networks, our rules make that clear".^[451][452] The lawsuit was dismissed by the United States District Court for the Northern District of California, upholding the Section 230 safe harbor, which dictates that the operators of an interactive computer service are not liable for the content published by its users.^[452][453] The lawsuit was revised in August 2016, providing comparisons to other telecommunications devices.^[450]

Twitter suspended multiple parody accounts that satirized Russian politics in May 2016, sparking protests and raising questions about where the company stands on freedom of speech.^[454] Following public outcry, Twitter restored the accounts the next day without explaining why the accounts had been suspended.^[455] The same day, Twitter, along with Facebook, Google, and Microsoft, jointly agreed to a European Union code of conduct obligating them to review "[the] majority of valid notifications for removal of illegal hate speech" posted on their services within 24 hours.^[456] In August 2016, Twitter stated that it had banned 235,000 accounts over the past six months, bringing the overall number of suspended accounts to 360,000 accounts in the past year, for violating policies banning use of the platform to promote extremism.^[457]

On May 10, 2019, Twitter announced that they suspended 166,513 accounts for promoting terrorism in the July–December 2018 period, stating there was a steady decrease in terrorist groups trying to use the platform owing to its "zero-tolerance policy enforcement". According to Vijaya Gadde, Legal, Policy and Trust and Safety Lead at Twitter, there was a reduction of 19% terror related tweets from the previous reporting period (January–June 2018).^{[458][459][460][461][462]}

As of July 30, 2020, Twitter will block URLs in tweets that point to external websites that contain malicious content (such as malware and phishing content) as well or hate speech, speech encouraging violence, terrorism, child sexual exploitation, breaches of privacy, and other similar content that is already banned as part of the content of tweets on the site. Users that frequently point to such sites may have their accounts suspended. Twitter said this was to bring their policy in line to prevent users from bypassing their tweet content restrictions by simply linking to the banned content.^[463]

Following the onset of protests by Donald Trump's supporters across the US in January 2021, Twitter suspended more than 70,000 accounts, stating that they shared "harmful QAnon-associated content" at a large scale, and were "dedicated to the propagation of this conspiracy theory across the service". The rioters that broke into the U.S. Capitol included a large number of QAnon followers.^[464]

Malicious and fake accounts

Between January and late July 2017, Twitter had identified and shut down over 7,000 fake accounts created by Iranian influence operations.^[465]

In May 2018, in response to scrutiny over the misuse of Twitter by those seeking to maliciously influence elections, Twitter announced that it would partner with the nonprofit organization Ballotpedia to add special labels verifying the authenticity of political candidates running for election in the U.S.^[466][467]

In December 2019, Twitter removed 5,929 accounts for violating their manipulation policies. The company investigated and attributed these accounts to a single state-run information operation, which originated in Saudi Arabia. The accounts were reported to be a part of a larger group of 88,000 accounts engaged in spammy behavior. However, Twitter did not disclose all of them as some could possibly be legitimate accounts taken over through hacking.^[468]

In March 2021, Twitter suspended around 3,500 fake accounts that were running a campaign to influence the American audience, after the US intelligence officials concluded that the assassination of The Washington Post journalist Jamal Khashoggi was "approved" by the Saudi Crown Prince Mohammed bin Salman. These Saudi accounts were working in two languages, English and Arabic, to influence public opinion around the issue. Many accounts commented directly on the tweets of US-based media houses, including The Post, CNN, CBS News and The Los Angeles Times. Twitter was unable to identify the source of the influence campaign.^[469]

As of 2022, the top four countries spreading state-linked Twitter misinformation are Russia, China, Iran and Saudi Arabia.^[470]

X bots

An X bot is a computer program that can automatically post, repost, and follow other accounts. X's open application programming interface and the availability of cloud servers make it possible for X bots to exist within the social networking site.^[471] Benign X bots may generate creative content and relevant product updates whereas malicious bots can make unpopular people seem popular, push irrelevant products on users and spread misinformation, spam or slander.^[472] Bots amass significant influence and have been noted to sway elections, influence the stock market, public appeal, and attack governments.^[473] As of 2013, Twitter said there were 20 million fake accounts on Twitter, representing less than 5% of active users.^[474] A 2020 estimate put the figure at 15% of all accounts or around 48 million accounts.^[11]

Usage

Dorsey (left) said after a Twitter Town Hall with Barack Obama held in July 2011, that Twitter received over 110,000 #AskObama tweets.^[475]

Impact

World leaders

Donald Trump's Twitter post

World leaders and their diplomats have taken note of Twitter's rapid expansion and have been increasingly utilizing Twitter diplomacy, the use of Twitter to engage with foreign publics and their own citizens. US Ambassador to Russia, Michael A. McFaul has been attributed as a pioneer of international Twitter diplomacy. He used Twitter after becoming ambassador in 2011, posting in English and Russian.^[527] On October 24, 2014, Queen Elizabeth II sent her first tweet to mark the opening of the London Science Museum's Information Age exhibition.^[528] A 2013 study by website Twiplomacy found that 153 of the 193 countries represented at the United Nations had established government Twitter accounts.^[529] The same study also found that those accounts amounted to 505 Twitter handles used by world leaders and their foreign ministers, with their tweets able to reach a combined audience of over 106 million followers.^[529]

According to an analysis of accounts, the heads of state of 125 countries and 139 other leading politicians have Twitter accounts that have between them sent more than 350,000 tweets and have almost 52 million followers. However, only 30 of these do their own tweeting, more than 80 do not subscribe to other politicians and many do not follow any accounts.^[530]

Censorship and moderation

Twitter is banned completely in Russia,^[534] Iran, China and North Korea,^[535] and has been intermittently blocked in numerous countries including Egypt, Iraq, Nigeria, Turkey, Venezuela and Turkmenistan on different bases.^{[536][537][538][539][540][541]} In 2016, Twitter cooperated with the Israeli government to remove certain content originating outside Israel from tweets seen in Israel.^[542] In the 11th biannual transparency report published on September 19, 2017, Twitter said that Turkey was the first among countries where about 90 percent of removal requests came from, followed by Russia, France and Germany.^[543] Twitter stated that between July 1 and December 31, 2018, "We received legal demands relating to 27,283 accounts from 47 different countries, including Bulgaria, Kyrgyzstan, Macedonia, and Slovenia for the first time."^[544] As part of evidence to a U.S. Senate Enquiry, the company admitted that their systems "detected and hid" several hundred thousand tweets relating to the 2016 Democratic National Committee email leak.^[545] During the curfew in Jammu and Kashmir after revocation of its autonomous status on August 5, 2019, the Indian government approached Twitter to block accounts accused of spreading anti-India content;^[546] by October 25, nearly one million tweets had been removed as a result.^[547]

In March 2022, shortly after Russia's censorship of Twitter, a Tor onion service link was created by the platform to allow people to access the website, even in countries with heavy Internet censorship.^[548][549]

Community Notes

Community Notes logo

Court cases and lawsuits

Twitter Inc. v. Taamneh, alongside Gonzalez v. Google, were heard by the United States Supreme Court during its 2022–2023 term. Both cases dealt with Internet content providers and whether they are liable for terrorism-related information posted by their users. In the case of Twitter v. Taamneh, the case asked if Twitter and other social media services are liable for user-generated terrorism content under the Antiterrorism and Effective Death Penalty Act of 1996 and are beyond their Section 230 protections. The court ruled in May 2023 that the charges brought against Twitter and other companies were not permissible under the Antiterrorism Act, and did not address the Section 230 question. This decision also supported the Court's per curiam decision in Gonzalez returning that case to the lower court for review in light of the Twitter decision.^[564][565]

In 2016, Twitter shareholder Doris Shenwick filed a lawsuit against Twitter, Inc., claiming executives misled investors over the company's growth prospects.^[566] In 2021, Twitter agreed to pay $809.5 million to settle.^[566]

In May 2022, Twitter agreed to pay $150 million to settle a lawsuit started by the Department of Justice and the Federal Trade Commission. The lawsuit concerned Twitter's use of email addresses and phone numbers of Twitter users to target advertisements at them. The company also agreed to third-party audits of its data privacy program.^[567]

On November 3, 2022, on the eve of expected lay-offs, a group of Twitter employees based in San Francisco and Cambridge filed a lawsuit in the U.S. District Court in San Francisco. Naming five current or former workers as plaintiffs, the suit accused the company of violating federal and state laws that govern notice of employment termination.^[568] The federal law in question is the Work Adjustment and Re training Notification (WARN) Act, and the state law in question is California's state WARN Act.^[569]

On November 20, 2023, Twitter filed a lawsuit against Media Matters, a media watchdog group. The lawsuit alleges defamation by Media Matters following its publication of a report claiming that advertisements for major brands were displayed alongside posts promoting Adolf Hitler and the Nazi party.^[570]

User accounts with large follower base

As of 25 July 2023, the ten X accounts with the most followers were:

Top ten most-followed X accounts

Rank	Change [lower-alpha 4]	Account name	Owner	Followers (millions)	Activity	Country
1		@elonmusk	Elon Musk	167	Business magnate and Chairman	ZAF  CAN  USA
2		@BarackObama	Barack Obama	131.9	44th U.S. president	USA
3		@justinbieber	Justin Bieber	111.7	Musician	CAN
4		@Cristiano	Cristiano Ronaldo	109.4	Footballer	PRT
5		@rihanna	Rihanna	108.2	Musician and businesswoman	BRB
6		@katyperry	Katy Perry	107.1	Musician	USA
7		@taylorswift13	Taylor Swift	94.2	Musician	USA
8		@narendramodi	Narendra Modi	91.4	Prime Minister of India	IND
9		@realDonaldTrump	Donald Trump	87.2	45th U.S. president	USA
10		@ladygaga	Lady Gaga	83.9	Musician and actress	USA

Record tweets

A selfie orchestrated by 86th Academy Awards host Ellen DeGeneres during the March 2, 2014, broadcast^[571] was at the time the most retweeted image ever.^[572] DeGeneres said she wanted to pay homage to Meryl Streep's record 17 Oscar nominations by setting a new record with her, and invited other Oscar celebrities to join them. The resulting photo of twelve celebrities broke the previous retweet record within forty minutes, and was retweeted over 1.8 million times in the first hour.^{[573][574][575]} By the end of the ceremony it had been retweeted over 2 million times.^[573] On May 9, 2017, Ellen's record was broken by Carter Wilkerson (@carterjwm) by collecting nearly 3.5 million retweets in a little over a month.^[576] This record was broken when Yusaku Maezawa announced a giveaway on Twitter in January 2019, accumulating 4.4 million retweets. A similar tweet he made in December 2019 was retweeted 3.8 million times.^[577]

The most tweeted moment in the history of Twitter occurred on August 2, 2013; during a Japanese television airing of the Studio Ghibli film Castle in the Sky, fans simultaneously tweeted the word balse (バルス)—the incantation for a destruction spell used during its climax, after it was uttered in the film. There was a global peak of 143,199 tweets in one second, beating the previous record of 33,388.^[578][579]

The most discussed event in Twitter history occurred on October 24, 2015; the hashtag ("#ALDubEBTamangPanahon") for Tamang Panahon, a live special episode of the Filipino variety show Eat Bulaga! at the Philippine Arena, centering on its popular on-air couple AlDub, attracted 41 million tweets.^[580][581] The most-discussed sporting event in Twitter history was the 2014 FIFA World Cup semi-final between Brazil and Germany on July 8, 2014.^[582]

According to Guinness World Records, the fastest pace to a million followers was set by actor Robert Downey Jr. in 23 hours and 22 minutes in April 2014.^[583] This record was later broken by Caitlyn Jenner, who joined the site on June 1, 2015, and amassed a million followers in just 4 hours and 3 minutes.^[584]