Write a reflection about the attachement reading.Twitter and the Paradox of Following and Trending
The Culture of Connectivity: A Critical History of
Social Media
Jose van Dijck
Print publication date: 2013
Print ISBN-13: 9780199970773
Published to Oxford Scholarship Online: January 2013
DOI: 10.1093/acprof:oso/9780199970773.001.0001
Twitter and the Paradox of Following and
José van Dijck
Abstract and Keywords
The fourth chapter traces the transformation of Twitter, the microblogging
platform that started in 2006. The platform aimed to be an autonomous utility
service promoting user connectedness, but gradually transmuted into an
information network exploiting user connectivity. Twitter’s history revolves
around a double paradox: first, the functions of following and trending presume
a neutral technological infrastructure where all users are equal and all content
is carried indiscriminately. In practice, Twitter’s filtering mechanisms inscribe
more weight to some twitterers and tweets, thus promoting the creation of big
followings and popular trends. Second, Twitter presents its network as an online
“town hall” for networked communication, but the platform has manifested itself
as a potent instrument for manipulating opinions. In light of this paradox, we
need to interpret how Twitter changed its initial ambitions from wanting to be a
“utility” to becoming an “information networking company.” Using instruments
like predictive analytics, the site increasingly aims at capitalizing the flow of
tweets rushing though its veins
Keywords: Twitter, microblogging, predictive analytics, affective economics, following, trending
4.1. Introduction
I think Twitter is a success for us when people stop talking about it, when
we stop doing these panels and people just use it as a utility, use it like
electricity. It fades into the background, something that’s just a part of
Page 1 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
communication. We put it on the same level as any communication device.
So, e-mail, SMS, phone. That’s where we want to be.1
When cofounder and executive Jack Dorsey prophesied Twitter’s prospects at a
Future of the Media panel in New York in 2009, he underlined that the company
wants users and developers to shape the platform into a generic infrastructure
for online communication and social interaction. Twitter first emerged in 2006,
at a time when nobody really knew what microblogging was. Early adopters had
experimented with several usages when Twitter really took off at a South by
Southwest conference in 2007. Six years after its launch, the platform has
become the world’s leading microblogging service, attracting almost 500 million
registered users and 88 million active users per month.2 Over the past half
decade, “tweeting” has sported multiple meanings, from sending instant short
messages to creating a live stream of instant opinion. Twitter’s technological
design was modified several times, and while establishing its brand name, the
platform experimented with various business models and governance strategies
to turn connectivity into a source of sustainable income. Social constructivists
would consider the platform’s development to be a stage of “interpretive
flexibility”: a stage (p.69) when a technology is still in flux and various,
sometimes contradictory, interpretations are wagered before stabilization is
reached (Bijker 1995; Pinch and Bijker 1984; Doherty, Coombs and Loan-Clarke
Dorsey’s stated objective of turning the platform into a utility, like water from
the tap or electricity from an outlet, entails a peculiar paradox. The terminology
presumes Twitter to be a neutral platform upon which users freely interact,
much like the Web itself—an infrastructure that transports streams of tweets,
regardless of who its users are and indifferent to the contents they exchange.
According to Twitter’s founding CEO, the infrastructure itself needs to fade into
the background, the way water pipes are made invisible and electricity cables
are taken for granted. From such a perspective, Twitter presents itself as an
echo chamber of random chatter, the online underbelly of mass opinions where
collective emotions are formed and where quick-lived trends wax and wane in
the public eye. And yet Twitter’s pipelines do not just transport streams of live
tweets; neither the platform nor its users are simple carriers of information.
Much rather, streams of data are engineered to promote certain uses and users
over others. Twitter presents the platform as a carrier, like a phone company,
but this objective is simultaneously challenged by the pressure to make its
content streams profitable. The paradox of enabling connectedness while
engineering connectivity, of propagating neutrality while securing profitability, is
played out in every aspect of the platform.
If we apply the concept of interpretive flexibility to Twitter’s first six years, we
can observe how conflicting meanings and divergent pressures informed the
microsystem’s development. The previous chapter described how Facebook
Page 2 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
gradually imputed expansive meanings to the online activities of “sharing,”
“liking,” and “friending”; in the case of Twitter, a similar evolution took place
with regards to social practices now known as “following” and “trending.”
Specific algorithms that inform these practices are presented as neutral, but in
fact apply filtering mechanisms to weigh and select user contributions and tweet
content. On the level of platform organization, Twitter had to walk a tightrope
between its ambition to be an autonomous communication network and the
commercial pressure to become an applied service accommodating advertisers.
To understand the implications of this ambivalence, we also need to track how
microblogging’s development plays out in a complex Internet milieu of push-andpull forces. How did Twitter position itself vis-à-vis its competitors in other
niches? What are the implications of the following and trending logic beyond the
platform itself? Posing the existential, strategic, and ecological question, this
chapter probes the validity of the platform’s aspiration to become a neutral
(p.70) 4.2. Asking the Existential Question: What Is Twitter?
When Twitter first arrived on the scene, nobody really knew how to define it.
Touted as the “SMS of the Internet,” the technology allowing users to send and
receive text-based messages of up to 140 characters known as tweets was
characterized as something between a short message service, a phone call, an email, and a blog: less cumbersome than keeping a blog, less exclusive than
talking to one person on a phone, less formal than e-mail exchange, and less
elaborate than most social network sites. The initial idea was for Twitter to be “a
sort of adrenalized Facebook, with friends communicating with friends in short
bursts.”3 The 140-character limitation was chosen not only because of its
conciseness and intensity, as we will see later on, but primarily for its technical
compatibility with mobile phone SMS services. The application quickly spread
via a number of other devices, such as PDAs, laptops, and desktops. Twitter’s
strength was its hardware versatility, as well as its capability of fitting multiple
online environments.4
If we take Jack Dorsey’s words seriously, Twitter was designed to be a
multipurpose tool upon which apps could be built. During the first years after its
emergence, Twitter was often called a service in search of a user application.
Predictably, users and markets are always looking for one specified function to
designate its value. Researchers, app developers, and journalists all tried to help
answer the existential question of what Twitter really was. Journalists wondered
about the technology’s most evident usage, let alone its “killer app” (Arcenaux
and Schmitz Weiss 2010). Information scientists “followed the hardware” to
understand the motives of early adopters, who are commonly eager to tweak
technologies to suit their needs or who invent needs for unspecified tools.5
Several researchers of information and communication technology attempted to
characterize Twitter by analyzing its activity streams (Krishnamurthy, Gill, and
Page 3 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
Arlitt 2008); others tried to define Twitter’s user rationale by mapping network
nodes in geographical space (Java et al. 2007). Researchers keenly observed that
subtle hardware adjustments affected interaction between (groups of) users, and
monitored the site’s architectural design to suggest modifications that would
strengthen the platform’s collaborative functions.6
Twitter rapidly managed to occupy a social networking niche that Facebook and
Myspace did not really serve. Over the years, a number of competing
microblogging platforms carved out niches within niches.7 Some of these
services were country-specific, and some combined microblogging with other
services, for instance file sharing. In contrast to its competitors, Twitter
positioned itself as an autonomous brand, unconnected to one (p.71) specific
tool, one specific country, or one specific paired service. For one thing, Twitter
steered its technological design to favor ubiquitous integrated use of its basic
architectural elements; in order to interlink with as many social networks as
possible, the microblogging service gradually adapted its hardware and tweaked
its software to fit other services’ standards. About the same time Twitter
emerged, Facebook added its own microblogging tool to its already hugely
popular site: News Feed highlighted information on recent profile changes,
upcoming events and birthdays, and other updates. In 2007, Facebook also
adopted Twitter on its site, which tremendously boosted the latter’s popularity.
Less than a year later, virtually every SNS provided links to Twitter, as did most
major news and entertainment organizations, ensuring it a winner-takes-all
position in this particular segment of the ecosystem.
So which distinctive features did Twitter’s interface integrate over time? For
starters, the platform firmly manifested itself as a user-centered site, a concept
cemented in the idea of following: users may subscribe to other users’ tweets,
and subscribers are known as followers. “Following” in the early years meant
engaging in a real-time communal dialogue, looking at and responding to
comments of users you were interested in. In a very short period, the platform
gained a critical mass of users, who wanted to engage in public or community
debates, exchanging suggestions and opinions. The Twitter logo transmogrified
into a click-on T-button that established “twittering” or “tweeting” as a
ubiquitous online practice, much as Facebook’s F-button did for “friending” and
the thumbs-up sign for “liking.” In line with Jack Dorsey’s ambition, the lightblue and white T-button arguably became the kind of turn-on switch you expect
to find in any social meeting place.
A significant expansion of Twitter’s architecture was the implementation, in late
2008, of “trending topics”—a feature that enabled users to group posts together
by topic by articulating certain words or phrases prefixed with a hashtag sign
(#). From now on, users could actively trend certain topics or passively track
topics. All trending topics are instantly indexed and filtered, before they become
visible in the “trending” sidebar. The Retweet function was rolled out in 2009;
Page 4 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
retweeting—users reposting interesting tweets from their friends using the
letters RT followed by the @ sign and a user name—became a very popular
function, generating enormous amounts of traffic on Twitter. By implementing
this feature, Twitter encouraged conversational tagging. Users started deploying
the tag itself as a prompt to create typically short-lived, heavily retweeted topics
(Huang, Thornton, and Efthimidias 2010). An important part of the push for
integrated use was the promotion of Twitter’s microsyntax as a new sort of (p.
72) currency. The symbols @ (referring to an online name) and # (denoting a
searchable topic) and RT (Retweet) were quickly absorbed into everyday
communication. The gradual appropriation of Twitter syntax across the Web and
beyond—on T-shirts and magnets—signals its ubiquitous integration in online
and offline social practices around the globe.
Most of the features mentioned so far emphasize Twitter’s purpose as a generic
service, developing its own techno-grammar to facilitate communication of a
specific nature across platforms. Starting in 2010, twitter.com rolled out its
revamped interface and presented itself as the “New Twitter,” a transformation
that confronted users with the site’s other (corporate) ambitions. Users could
now click on individual tweets to see pictures and video clips from a variety of
other websites, including Flickr and YouTube, without leaving Twitter. Like
Facebook and Google, Twitter started to require external developers comply
with the OAuth standards. The site also inserted geospatial features, allowing
users to initiate conversational interactions in a specific location or area and
track them. Besides, Twitter introduced Promoted Tweets and Promoted Trends
—a trending topic or widespread tweet paid for by a sponsor. Evidently, these
new features promoted frictionless sharing between platforms and expanded
Twitter’s commercial potential, as they facilitated the introduction of new
business models—an element I will return to in the next section.
An even more thorough interface redesign, also known as the “New-new
Twitter,” came in late 2011, shortly after Facebook had announced its Timeline
feature. This revamping introduced four new buttons: the Home button, showing
a “timeline” of tweets of people you follow; a Connect button, symbolized by the
@ sign, displaying who and what you are following and retweeting; a Discover
function, shaped like #, which forwards “the most relevant stories from your
world”; and the Me button, a feature that contains your profile, direct messages,
and favorites.8 Twitter’s new interface layout structured users’ navigation while
also rendering input more uniform and thus more accessible to advertisers.
Much like Facebook’s narrative Timeline structure, Twitter’s New-new interface
experience comes across as more streamlined than the old one. The choice of a
standardized design is obviously a response to major competitors’ interface
strategies, not just Facebook but also Google+.
Page 5 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
Improving the potential for interoperability and frictionless sharing across
platforms, Twitter’s interface overhaul reflects an attempt to weave its
idiosyncratic microsyntax into the fabric of sociality: hashtags, RTs, and @replies
moved to the center of each member’s online experience. If Twitter were up to
now largely regarded a site for technology-savvy people, the New-new Twitter
helped push the site’s image as a common utensil for (p.73) ordinary users—
one step closer to becoming a “utility.” The platform’s functionality as a network
that helps users to connect, and to initiate and follow conversations worldwide,
obviously generated a mass of tweets and twitterers. The gradual rollout of
coding and design changes reveals a pattern where features enhancing user
connectedness are gradually complemented by features advancing exploitation
of the site’s connectivity.
Users and Usage
From the onset, users and governments embraced Twitter as a tool for
connecting individuals and communities of users—a platform that empowers
citizens to voice opinions and emotions, that helps stage public dialogues, and
supports groups or ideas to garner attention. In August 2010, Wikipedia listed
nine “notable uses” for Twitter, each describing a real-life (or real-time) context
in which Twitter had recently functioned as a central tool: in campaigning, legal
proceedings, education, emergencies, protest and politics, public relations,
reporting dissent, space exploration, and opinion polling.9 One can read this list
of notable uses as a haphazard inventory of the various social contexts in which
Twitter penetrated people’s everyday lives. Wikipedia’s list singles out social
circumstances in which users appropriated the tool, but that does not imply the
tool itself is not a shaping force. If we examine the mutual shaping between
Twitter and its users in specific social settings, we can hardly escape the
ambiguous proposition inscribed in the platform: while claiming to facilitate all
voices evenly, at the same time the site crafts a hierarchical structure of
twitterers. I will zoom in on two specific usages to illustrate this paradox:
Twitter as a tool for (community or political) organizing and as a platform for
Twitter users tend to be perceived as a mass of mostly young active users who
feel empowered by the service in their individual contributions to dialogues or in
their collective efforts to influence public debates. The platform’s demographics
and user dynamics reveal a more nuanced picture, though. In the early years of
its existence, Twitter’s user demographics stood out from other SNSs. Most
social media, such as Facebook and Myspace, gained their popularity from large
contingents of young, educated users—teenagers, college students, young
professionals—who, at that stage in their lives, are eager to establish as many
contacts as possible, at both a personal and a professional level. In contrast to
competing SNSs, Twitter’s initial user base largely consisted of older adults,
many of whom had not used other social sites before. For the first couple of
years, the (p.74) microblogging network attracted users mostly in business
Page 6 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
settings and news outlets, resulting in an early adopter profile of older (35 and
up) professional users.10 Once the Twitter audience started to soar, markedly
after May 2009, the group of younger adults grew at a much faster rate,
resulting in a user constituency aged predominantly 35 and under.11 As Twitter
began to filter into the mainstream by catering news feeds and celebrity
updates, the swing coincided with a shift in gender demographics.12
Along with the demographic changes came gravitation in user dynamics toward
a few heavy twitterers: a small but prolific group of 10 percent of Twitter users
accounts for over 90 percent of tweets (Heil and Piskorski 2009).13 In 2010,
information scientists reported from their comprehensive data analysis of all
Twitter users that only 22 percent have reciprocal relationships, while 68
percent of users are not followed by any of their followings: “People follow
others not only for social networking but for information, as the act of following
represents the desire to receives [sic] all tweets by the person” (Kwak et al.
2010: 594). The popular RT and # buttons were exploited by few users to
become influential twitterers. However, the most influential positions were not
exclusively reserved for users who spawn large numbers of retweets; indeed, the
most skillful users can exert substantial influence over a variety of topics, as
researchers conclude: “Influence is not gained spontaneously or accidentally, but
through concerted effort such as limiting tweets to a single topic” (Cha et al.
2010: 10).
Despite Twitter’s image as the online “town hall” for networked communication
—a mere amplifier of individual voices as well as collective opinions—the
platform manifested itself increasingly as a potent instrument for routing ideas
and manipulating opinions. On a self-proclaimed neutral “utility-like” platform
such as Twitter, one would expect all users to be equal. But some users are more
equal than others—an inequity that is partly due to the platform’s architecture,
as described above, and partly to users’ own active steering. The ideal of an
open and free twitterverse in reality comes closer to a public dialogue ruled by a
small number of hyperconnected influencers skilled at handling large numbers
of followers. The platform’s architecture privileges certain influential users who
can increase tweet volume, and whom thus garner more followers. Twitter’s
ambition to be an echo chamber of serendipitous chatter thus finds itself at odds
with the implicit capacity, inscribed in its engine, to allow some users to exert
extraordinary influence. How did this ambiguity manifest itself in specific social
The first social context in which Twitter’s dilemma surfaced is that of political
(grassroots) activism. The Iranian uprising, in 2009, the Arab Spring in 2010,
and the Occupy movement of 2011 were all considered (p.75) examples of
users’ empowerment through social media—citizen’s taking hold of their own
communication and propaganda channels to challenge the power of conventional
gatekeepers such as governments and news organizations. The 2009 revolt in
Page 7 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
Iran was hailed as the “Twitter revolution,” stressing a view of social media as
inherently liberating (Gaffney 2010; Giroux 2009). Foreign media and
government officials attributed major significance to social media platforms in
the Arab Spring uprisings; they considered them neutral tools, which, in the
hands of prodemocratic citizens, rendered them powerful as a collective.
Detractors such as Morozov (2011) and Christensen (2011) have raised serious
doubts about social media’s liberating potential, arguing that most platforms
aggravate the grip of totalitarian regimes on protesters because actual users can
be easily tracked through these tools.14
According to critics of the liberating view, Twitter is not a neutral platform, and
all users are not equally influential. Twitter’s filtering functions are constantly
fine-tuned, gauging the influence of its users in order to better organize search
results (Huberman, Romero, and Wu 2009). Twitterers like CNNs Christiane
Amanpour or some Middle East specialists are automatically assigned more
weight in the twitterverse than just anyone who offers an opinion. On top of
these filtering mechanisms, the voices of protestors using Twitter during the
uprisings were not all equally influential. In their thorough analysis of the
protests in Tunisia, Poell and Darmoni (2012) show that twitterers at the center
of the Arab diaspora networks, who were well connected with activist
organizations and journalists around the world, were able to use the platform for
distributing messages to international media. Just as EdgeRank filters Facebook
“friends” for relevance, Twitter’s algorithms, policies, and user practices consign
different weight to different voices.
A second context in which to demonstrate the platform’s ambiguity and its users’
differentiated authority is Twitter’s role as a tool for self-promotion. While
Twitter boasts of its capability to resonate the “roar of the crowd,” the platform
savors the stardom of influential twitterers, especially of stars, politicians, and
celebrities. Famous personalities profit from Twitter’s followers’ function as the
perfect medium to organize and maintain their fans. For politicians, Twitter has
become an indispensable tool with which to galvanize their electoral base
because the medium allows them to control their messages—a big advantage
over mainstream media, where they are dependent on journalists’ framing. Not
surprisingly, celebrities and politicians top the list of most frequently followed
twitterers, resulting in a ranking where the top five attracts millions of
followers.15 Most star twitterers employ professional PR staff to manage their
devotees, (p.76) voters, or fans. Twitter, for its part, supplies consultants to
keep the high-ranking influencers on top; after all, the traffic volume these stars
create is quite lucrative to the company in terms of selling advertising space—a
strategy I will return to in the next section.
Nursing a big following is not just a celebrity’s privilege, however; for a lot of
average users, Twitter has become a prime tool to advertise the self. The sheer
number of followers has become a barometer for measuring popularity and
Page 8 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
influence, ascribing more power to few users in the twitterverse. Individuals
quickly learned how to play the system and accumulate a lot of clout on Twitter.
According to New York magazine journalist Hagan, the “impulse to make life a
publicly annotated experience has blurred the distinction between advertising
and self-expression, marketing and identity.”16 Users want large followings not
just for reasons of vanity or self-esteem: they may actually cash in on their
popularity rating by selling their influence to the highest bidder. Specialized
sites such as Klout automatically calculate an individual’s influence on the Web,
using an algorithm that is largely based on Twitter followings and Facebook
connections, and then sells this information to potential advertisers or
companies looking for online influencers.17 Companies and advertisers are keen
to pay powerful twitterers—both celebrities and ordinary users—to distribute
their brand name.
As these two social uses of Twitter illustrate, explicit users are shaping the
platform’s direction at the same time, and by the same means, as the platform is
shaping users’ behavior and social standing. The meaning of “following” initially
was connecting to someone for the purpose of interaction and exchange.
Gradually the term also came to signify trailing someone and “buying into” his
or her ideas. To examine the equally ambiguous meaning of “trending,” we will
now look more closely at the content and cultural form of tweets.
Content and Cultural Form
Both the quality and quantity of tweets have been vital elements in the dispute
of what constitutes Twitter’s essence during the stage of interpretative
flexibility. The “tweet” is arguably Twitter’s most distinctive contribution to
online culture; a sentence limited to 140 characters flagged by a hashtag has
become a global format for online public commentary. This new cultural form
has been adopted widely outside the platform proper, for instance in newspapers
and on television. Its concise syntax and delimited length render the tweet
virtually synonymous with a quote—a citation from a source for which
authentication resides with the platform, not the journalist. Aside from figuring
in the news, the tweet has emerged as a cultural (p.77) form inspiring poets
and literary authors. So-called “twitlit” is yet another sign of Twitter’s
microsyntax becoming part of globalized cultural discourse.
The quality of tweets has been subject to interpretive contestation; most
discussions revolve around the question whether tweets are conversational or
informational, and whether they contain essential or nonessential information.
Some early critics touted the flow of tweets as verbal diarrhea, while others
characterized it, in line with Twitter’s ambitions, as a gush of free-floating public
chatter.18 During the first three years of Twitter’s development researchers
discussed whether the tool just supported everyday small talk or whether it had
news-signaling significance.19 Indeed, Tweets may be about the latest
development in the Middle East or about Lady Gaga’s cold.20 But essential and
Page 9 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
nonessential content have always coexisted in older media, even in what we call
quality newspapers; it is hardly surprising, then, to find these two types of
content emerge side by side in a new medium. Perhaps most typical of a Tweet is
not the “weight” of its content, but the message’s expressive and personal
nature. Researchers Marwick and boyd (2011) observe that tweets with the
pronoun “I” in them tend to be well received. Politicians use tweets to wrap their
political messages in personal narratives in order to exploit the intimate
connotation of the medium’s communicative mode. Tweets also work best to
convey affective content, both in terms of gut-fired opinion and spontaneous
reactions. Increasingly, politicians’ and celebrities’ “personal” tweets appear in
the mass media as commentaries, replacing the perfunctory quote. Fitting in
with Facebook’s narrative strategy, the tweet’s effectiveness lies in having a
personalized public message enter a customized online social environment.
The most vital qualifier of the paradoxical nature of tweets, though, lies not in
their quality but in their enormous quantity. The daily number of Twitter
messages increased from 27 million in 2009 to 290 million in February 2012.21
On the one hand, tweets flowing in real time are often conceptualized as a
stream of (global) consciousness or a resonance room for unmediated public
chatter. On the other hand, the concept of an indiscriminate flow of tweets is at
odds with Twitter’s efforts to structure the amorphous information stream into a
usable, exploitable resource. Not every tweet is assigned the same importance:
as described above, some senders are valued over others, and some tweets
weigh more than others. Weight is measured in tweets per second (TPS): when
intensity is high, they are assigned more impact. The algorithmic push of
intensity over quality results in short periods of heavily circulated messages that
may become trends. Trending topics may thus refer to streams of “surfacing
content” but may also signal content-massages aimed at pushing the message to
go viral and spill over into other social platforms and mainstream media. The (p.
78) double logic of “tracking emerging trends” and “setting a trend” poses a
profound challenge to Twitter’s owners. As one of the platform’s CEOs, Dick
Costello, rightly observed, the “danger of overstructuring information is that the
user stops experiencing Twitter the way people originally came to experience
Twitter as the place for free-form, serendipitous chatter.”22
Twitter’s investment in trending topics could be regarded emblematic for the
platform’s divergent ambitions to simultaneously measure, engineer, and
mobilize the public’s mood. Both the platform’s coding technologies and its
users help content surface while, vice versa, trending topics help certain users
to become influencers. The function of a platform as a global opinion machine is
hardly novel. Jürgen Habermas (1989) theorized that media channels, far from
registering the free exchange of ideas, constitute a social space through which
norms for communication and interaction are produced. Referring to
newspapers and television news, Habermas argued that by using publicity
strategies such as opinion polls, mass media infiltrated the formal public sphere,
Page 10 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
where they “corrupted” collective opinion formation by introducing mechanisms
that served corporate or government interests (van Dijck 2012a). In other words,
opinion polling also includes opinion engineering.
What is new in microblogging is that the tweet flow, in contrast to the
programmed television flow, is conceptualized as a live stream of uninhibited,
unedited, instant, short, and short-lived reactions—a stream that supposedly
taps a real-time undercurrent of opinions and gut feelings. If we accept at face
value the owners’ claim that Twitter is a neutral infrastructure where the
current gist of tweets gushes unmediated through its pipelines, this ambition is
controverted by the fact that the system is engineered to manage the flow and
gets manipulated by hyperconnected users who want to influence the flow. To
understand the inherent paradox of Twitter’s following and trending logic, it
may not be enough to ask the question of what Twitter is; we also need to raise
the question of what Twitter wants, meaning we have to look at its
socioeconomic structure. How does the company envision its corporate strategy,
including its governance policy and (future) business model?
4.3. Asking the Strategic Question: What Does Twitter Want?
As stated in the previous section, Twitter’s owners wanted to develop the
platform as an independent microblogging service. Like Facebook, Twitter
concentrated on making its service compatible with as many platforms as (p.79)
possible, securing the ubiquitous presence of its iconized light-blue and white Tbutton in all kinds of hardware (mobile phones), software (websites), and media
content (TV programs, ads, newspapers). Facebook and Google have integrated
Twitter in their services—likely at the expense of their own microblogging
services News Feed and Buzz—because Twitter’s brand name quickly attracted
lots of traffic. Twitter’s status as a stand-alone company, though, has been a
regular subject of speculation; market analysts keep a constant eye on the site’s
monetizing strategy, whether as an autonomous brand or as a potential takeover
target for other platforms.23 So far, the company has proved able to retain its
sovereign status, ostensibly anchoring its claim to neutrality in independent
ownership status, but how long will the company manage to live up to its
original ideals? And, learning from Facebook’s IPO, would a future public
offering compromise the owner’s objectives and thwart users’ enthusiasm?
One of Twitter’s problems in retaining independence is related to the company’s
ambiguous ambitions: can Twitter be a utility that facilitates connectedness
while at the same time being a company that exploits the giant stream of tweets
and metadata its users generate? The answer is simple: it needs to do both if it
wants to survive. Twitter’s position in the ecosystem has always been a
precarious one. Its rivals, Facebook and Google, each dominate a demarcated
territory in the ecosystem: Google “owns” search, Facebook “owns” social
networking. Sure, Twitter “owns” microblogging, but its niche is not (yet) as
Page 11 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
clearly branded as that of its competitors. In 2011, CEO Dick Costello began to
call Twitter an “information network” rather than a “social networking service”
to signal the company’s move toward the connectivity approach. Shying away
from Jack Dorsey’s earlier “utility” ambitions and attaching a new generic label
to the platform may well be interpreted as a territorial defense strategy—a
deliberate attempt to wall off Twitter’s niche from its looming contenders’ turf.
“Information networking” could be considered the company’s revamped
corporate aim, still in the early stages of development—microblogging’s period
of interpretive flexibility being far from over. However, new corporate goals
require strategic positioning, not only toward third-party developers that
historically popularized Twitter by designing a large variety of apps, but also
toward “old” media building on Twitter traffic mechanisms and, last but not
least, toward big social media competitors.
Until 2010, the company never charged for the use of its data, and many app
designers proved to be much better than Twitter itself in monetizing their daily
gush of tweets. For instance, TweetDeck, a popular monitoring tool that used the
Twitter API to form a sort of dashboard for trailing and redirecting online traffic,
became a successful start-up. In 2011, Twitter (p.80) took over TweetDeck,
thus underscoring its new ambition to monetize content rather than exploit a
utility. From the side of “old” media, television producers increasingly integrated
microblogging tools into their professional routines for making news or
entertainment. A number of newspapers signal “trending topics,” based on
Twitter’s most popular tweets of the day, and deploy Twitter’s potential to tap
into near-boiling topics to anticipate and create news.24 In addition, broadcast
producers exploit Twitter’s “viral looping” capacities by drawing audiences into
a conversation, while the company can also provide aggregated real-time
analytics that help one understand the dynamics of televised debate or detect
sentiment pulses in staged entertainment (Diakopoulos and Shamma 2010). If
Twitter wants to be a leading service in information networking, we should not
be surprised to see partnerships with major content producers and broadcast
industries emerge.
Twitter’s independent position as an “information network” seems far from
secure if we consider the importance of search in any setting that centers on
(meta)data interpretation and exploitation. And with the increased emphasis on
information and search comes a scenario in which Twitter has no choice but to
team up with the giant of search: Google. Twitter’s annexation by one of the big
content providers may seem unlikely at this point, but the above examples of
takeovers and alliances suggest that Twitter’s autonomy is anything but
guaranteed in the volatile ecosystem of connective media.25 If Twitter wants to
be an information company rather than a utility, this choice not only affects
Twitter’s ownership status but also its choice of business models.
Page 12 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
Business Models
Up until 2009, Twitter’s owners remained deliberately vague about plans to
monetize their popular service; they raised enough money from venture
capitalists to allow time to find a suitable revenue model.26 At some point,
though, business analysts began to ask whether Twitter’s owners were
interested in monetizing the service at all; just as media watchers initially called
Twitter a service in search of a user application, market analysts wondered
whether Twitter, four years after its launch, was still in search of a business
model.27 Like other social networking sites, notably YouTube and Facebook,
Twitter relied on the strategy of building an audience of users first and finding
revenue streams later. In these years, management experts were still convinced
that the largest possible user base is crucial to a site’s long-term sustained
profitability (Enders et al. 2008). However, (p.81) business models are not
ready-made strategies; testing business models, building a user base, and
tweaking the interface are mutually defining tactics—a dynamics that few
economists recognize (Orlikowski and Iacono 2001).
When Twitter still regarded itself as a general conduit for communication traffic,
it aimed at helping businesses create customer value. Hence, the company
appeared mostly interested in business models that involved generating service
fees from users or revenue fees from corporate clients. The first option soon
turned out to be illusory—users were not going to pay for a subscription if they
could use competing services for free. The second option looked more
promising. From the very beginning, Twitter encouraged third-party developers
to design APIs by opening up its metadata to everyone, from researchers to
commercial developers, without charge. Several years into the site’s existence, a
number of books already explained the tool’s power to “dominate your market”
or how to “get rich with Twitter” (Comm and Burge 2009; Prince 2010). The
company stood by and watched as outside programmers developed monetizing
services and tested them in the marketplace.28
For the first couple of years, Twitter also resisted the business model of paid
advertisements showing up next to the Twitter dialogue box. It was not until the
implementation of Twitter’s trending topics and geospatial features that the
company started to exploit data for targeted, personalized advertising. In the
spring and summer of 2010, some cautious steps in Twitter’s slow-rolling
business model could be observed. The site launched @earlybird Exclusive
Offers, offering followers time-sensitive deals on products and events from
sponsors, and introduced Promoted Tweets and Promoted Trends, a service
linking keywords to advertisers in order to insert sponsored tweets into the
stream of real-time conversation.29 Geospatial features, such as Points of
Interest, allowed for location-based targeted messages. Earlier that year, Twitter
had bought up Summize, a successful start-up that exploits search engines
linked to geo-location systems. By monetizing these new features, Twitter paved
Page 13 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
the way to include sponsored content—push-based, pull-based, or geo-based—
next to Twitter messages.
It took more time for the company to figure out how to utilize the increasingly
precious resource of connectivity flowing nonstop through the site’s veins,
especially the massive amounts of instant, spontaneous reactions to themed
tweets. Indeed, Twitter had already sold the rights to include tweets in search
results to Google and Microsoft in 2009, which turned out to be a lucrative deal.
And by starting to charge external developers for using Twitter’s data in 2010,
the company recouped some of that (p.82) value. Still, some Twitter watchers
and market analysts became impatient with the company’s slow learning curve
when it came to business models:
For those of you not familiar with the firehose, it’s the data that flows
through Twitter from its 175 million users and the 155 million daily tweets.
That’s a lot of valuable data that can be used for a variety of reasons by
Twitter and third parties. Currently, Twitter charges some companies to
access its data but most companies pay a modest amount or nothing to
access the data stream. But as data becomes more important, the value of
Twitter’s firehose becomes more valuable. It means that Twitter has a huge
revenue source under its nose, as long as Twitter is willing to capitalize on
In other words, if Twitter takes seriously its ambition to become an information
company, exploiting real-time user data is an indispensable means to this end.
Researchers tested Twitter’s effectiveness as a “social sensor” for real-time
events, such as accidents or earthquakes, which could be measured almost
instantly by tracking the tagged trends on Twitter (Sakaki, Okazaki, and Matsuo
2010). In addition, academics examined Twitter’s potential as an instrument for
“sentiment analysis” and “opinion mining” (Diakopoulos and Shamma 2010; Pak
and Paroubek 2010) and for measuring the public mood with the help of “realtime analytics” (Bollen, Mao, and Pepe 2010).
Building a business model on this potential invokes the paradox explored in the
previous section while also adding another one. Besides the tension between
perceptions of Twitter as a neutral platform while, in reality, its mechanisms
promote engineering and manipulation, the platform also faces the dilemma of
what Andrejevic (2011) calls “affective economics.” Sentiment analysis, mood
mining, and opinion mining—all subsets of predictive analytics—largely work on
the presumption that Twitter is indeed a barometer of emotion and opinion, but
these methods entail more than taking the “emotional pulse” of the Internet.
Complex algorithms derived from past user behavior are used to anticipate but
also affect future behavior: The process of aggregating and disaggregating data
from individual consumers may be deployed to tap into users’ buzz about brands
as well as to create brand communities based on Twitter dialogues on specific
Page 14 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
topics. As Andrejevic (2011) concludes: “Any consideration of affective
economics should include the ways in which marketers seek to manage
consumers through the collection not just of demographic information, but of
extensive real-time databases of their online behavior and conversations” (604).
While Twitter’s monetizing strategy is still open and ambivalent, the company
carefully pursues various business models, rolling out changes in (p.83)
interface gradually—a sort of controlled experimentation with different models
alongside each other (McGrath 2009). After all, every modification involves the
risk of losing users—a precious asset for an “information networking” company
that is completely dependent on users’ willingness to transmit content. As
Twitter develops its business models in close relation to coding technologies,
user routines, and content, we see that the company’s transforming ambitions
result in a double paradox. Twitter-the-ambient-utility promoting user
connectedness finds itself at odds with Twitter-the-information-network
exploiting connectivity to help businesses promote their brands among users.
Platform owners, users, and third parties are engaged in a precarious
choreography that surfaces most poignantly at the level of governance.
In contrast to Facebook, Twitter’s terms of service (ToS) have always been quite
forthright in stating the rights and restrictions of users versus owners and of
users versus third-party developers.31 When Twitter started in 2006, its terms of
service were very general and did not say much about the way users could
deploy the tool to communicate: “We encourage users to contribute their
creations to the public domain or consider progressive licensing terms.” Unlike
Facebook users, twitterers never fostered the illusion of privacy or intimate
circles, as they use the service largely as a tool for (self)-promotion and
interaction with the world at large. Twitter’s policy in 2009 explicitly articulated
the company’s intention to offer an open service to global communities: “What
you say on Twitter may be viewed all around the world instantly. You are what
you Tweet!” In other words, users should control their public statements, while
the platform vows to protect a users’ right to express their views and own their
content. As stated in the ToS: users “retain rights to any Content [they] submit,
post, display on or through Twitter.” Twitter users have never taken the site’s
owners to court for violation of privacy rules, and, to many users’ surprise, the
company indeed protected twitterers’ freedom of speech: they even challenged a
defeat in a subpoena case to hand over a specific user’s tweets by taking the
decision to a higher court.32
Users were a lot more critical, though, when Twitter announced its policy, in
January 2012, of complying with each country’s laws and censor tweets per
region, whereas previously Twitter would only block tweets or accounts globally.
From now on, a specific tweet could be prevented from appearing in a particular
country or region, replaced by a gray box with a (p.84) note reading, for
Page 15 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
instance, “This tweet from @username has been withheld in: Thailand.” A
spokesperson explained that Twitter had no choice but to abide by national laws
and comply with local legal requirements if the platform were to stay in service
in these countries. The example of German law prohibiting Nazi material to be
published on any online platform provided an example of the kind of legal
constraints the company faced. Yet Twitter users feared a censorship policy that
threatened First Amendment rights in many non-Western countries. A “Twitter
Black-out day” was staged on January 28, 2012, to protest this rule.33 Using the
hashtag #TwitterBlackout, the boycott rallied support among thousands of users
in various countries, gathering responses such as “Twitter starts deleting
tweets, I stop posting tweets. Join the #twitterblackout tomorrow!”
While most SNSs’ terms of service, particularly Facebook’s, tend to regulate
relationships between platform owners and users, Twitter deployed its
governance terms also to arbitrate the interests of its users vis-à-vis those of
third-party developers. The platform’s first ToS in 2006 said nothing about the
use of data by app developers or by the platform’s owners for advertising
purposes. In 2009, the ToS added that all data sent through Twitter may be used
by third parties: “You agree that this license includes the right for Twitter to
make such Content available to other companies, organizations or individuals
who partner with Twitter for the syndication, broadcast, distribution or
publication of such Content on other media and services, subject to our terms
and conditions for such Content use.” On the condition of the owner’s consent,
and later by paying a small fee, app developers were thus granted access to
Twitter data. The introduction of Promoted Tweets and Trends led to a change in
the 2011 terms of service, which now contain a line stating that “Twitter and its
third party providers and partners may place such advertising on the Services or
in connection with the display of Content or information from the Services
whether submitted by you or others.” Even if users were not very happy with
Twitter’s new advertising policies, they did not protest them either, apparently
accepting the new norm of letting a free service be paid for by ads.
However, the inclusion of a new clause in Twitter’s revised ToS policy in 2011,
aimed at anchoring the platform’s broadened options for data mining, raised
more eyebrows, especially those of third-party developers:
Twitter uses a variety of services hosted by third parties to help provide
our Services, such as hosting our various blogs and wikis, and to help us
understand the use of our Services, such as Google Analytics. These
services may collect information sent by your browser as part of a web
page request, such as cookies or your IP request. (Emphases added)
(p.85) If before 2011, all third parties were indiscriminately allowed to develop
Twitter services on the basis of its API, this clause grants this privilege to select
third parties on the condition that they comply with strict rules. The statement
Page 16 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
also reveals that Twitter partners with Google to develop data-mining strategies.
Instead of users lashing out at Twitter for selling out their data, third-party
developers protested what they saw as unprecedented restrictions on their
freedom to develop apps for a service once known as a “utility.” A company
spokesman defended Twitter’s new rules, citing the need for a more “consistent
user experience” across platforms, because “consumers continue to be confused
by the different ways that a fractured landscape of third-party Twitter clients
display tweets and let users interact with core Twitter functions.”34 He
confirmed that third-party developers could continue to create client
applications, such as content-curating or publishing tools, as long as they
adhered to Twitter’s ToS. Unsurprisingly, the announcement triggered furious
reactions among third-party developers, who felt excluded and rejected after
many years of helping the platform gain ubiquitous visibility in the vast
ecosystem of connective media. As one blogger expressed his dismay:
You [Twitter] may feel you “need” this consistency, but you don’t. You want
it, and are willing to make tradeoffs to get it. I just hope you realize how
big those tradeoffs are, and how chilling it is for Twitter to decide that only
certain kinds of innovation on the Twitter API are welcome.35
It is difficult not to read these ToS modifications as the next step in Twitter’s
new ambition to become an “information company.” While Twitter’s governance
policy explicitly protects users’ rights to express and control their opinions, the
company is forced to comply with country-specific legislation if it wants to stay
in business; and the company’s vocal defense of users against exploitation of
their data by third-party developers draw a thin veil over the company’s need to
team up with a data giant to help monetize Twitter’s vast resources.
Within the short time span of six years, the company’s ambitions transformed
from wanting to be a global, neutral communication channel for citizens that
might be used in defiance of governments, to wanting to be a profitable venture
that must obey the laws of those countries where it seeks to attract customers.
Twitter also risked its reputation as an “open” platform facilitating all potential
developers by restricting access to its precious tweet flow and seeking exclusive
collaboration with few powerful allies. A detailed exploration of this paradox
likely does not answer the question of what Twitter wants; it rather exposes the
minefield of conflicting forces in (p.86) which the company operates. And this
minefield of forces inevitably draws attention to the larger ecosystem of
connective media in which Twitter has evolved and is bound to play a role in the
4.4. Asking the Ecological Question: How Will Twitter Evolve?
The gradual transformation of Twitter as an autonomous utility promoting user
connectedness into an information network exploiting user connectivity can
hardly be evaluated in isolation. The double paradox laid out in the previous
Page 17 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
sections did not simply emanate from the owners’ decision to change objectives.
Instead, the proliferation of Twitter as a platform has been a complex process in
which technological adjustments are intricately intertwined with modifications
in user base, content channeling, choices for revenue models, and changes in
governance policies and ownership strategies. By tracking the process of
interpretive flexibility in its first six years, I sketched how Twitter developed as a
niche service in the social networking domain to one of the major online players.
And yet Twitter’s embeddedness in the larger ecosystem, which is partly
responsible for its past growth, will also account for some of its future
vulnerabilities. As of the spring of 2012, Twitter found itself at the crossroads of
a maze of interdependent platforms where it needed to define its position vis-àvis users as well as its various competitors.
Perhaps against all odds, Twitter is still an independent company, having
conquered a dominant position in a niche of online sociality it has itself coded
and branded. Microblogging has become virtually synonymous with tweeting.
The platform’s microsyntax of @ and # epitomize how following and trending—
both as active and as passive verbs—are part of a quotidian discourse the basic
grammar of which is understood even by nontwitterers. The brand is perceived
as a generic practice, much like “googling.” Twitter’s early image as an echo
chamber of serendipitous chatter, a town hall of public dialogue, and an
amplifier of suppressed voices still lingers in the minds of people.36 This image
motivates other social media platforms and traditional media to mobilize Twitter
as part of their armamentarium. One might well argue that Twitter’s future
success as an information network is conditional on the strong resonance of its
earlier ambition to be a neutral utility; many users still regard the platform’s
utility function as its prime objective and hence accept the platform’s claim to
neutrality even if the exploitation and manipulation of tweets has gained wider
recognition. In the face of the platform’s refurbished ambitions, the question is
how Twitter will live up to that new paradox-based image.
(p.87) Twitter’s strongest asset in the competition with other platforms is its
ability to generate enormous amounts of “live” streams of short-lived online
traffic that can be minutely tracked in real time. Neither Facebook nor Google+
nor YouTube is equipped to serve this function. Therefore, Twitter hopes to
strategically position itself in the market of predictive analytics and real-time
analytics. The algorithmic drive to analyze online behavior on the basis of past
and live behavioral data bolsters the cultural logic of following and trending as
practices that simultaneously comprise a passive reflection and active
manipulation of social motion. Amazon’s famous taste recommendation system is
just one example of the predictive analytics that is growing more ingenious and
powerful by the day; major offline and online companies hire engineers to
evaluate their customer’s data for individualized tracking and targeted
advertising.37 The new hot science of online data engineering interprets
consumer data while concurrently steering consumers’ desires. No competing
Page 18 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
platform has access to vast streams of “live” social data the way Twitter does,
but developing the tools to analyze these data streams and translate them into
profitable algorithms is another matter. With Google Analytics as its preferred
partner, Twitter may well strengthen its position as a global player in the
information-networking field, because “search” is a crucial factor in the
successful deployment of predictive and real-time analytics. The alliance with
Google also brings a vast potential for integrating online advertising services.
Besides playing a growing role in online consumer markets, trend mining and
opinion mining are also deployed for nonmarket purposes, varying from political
campaigns to promoting civilian grassroots causes. Twitter celebrities with large
followings (Oprah, Beyoncé) are regularly mobilized to bring noble causes to the
attention of their followers, whether poverty in Sudan or animal cruelty in
Puerto Rico. Hashtag activism, as it is also called, embraces Twitter as a tool for
instant propagation of social causes, and Twitter is a powerful weapon in the
hands of grassroots groups and NGOs like Greenpeace to harness consumer
awareness or organize boycotts. The flipside of low-threshold, Twitter-driven
protests is that they become shorthand for easily garnered public support.
Indeed, as some journalists observe, protests-by-the-click are now so common
that hashtag activism fatigue threatens to blur all causes.38 More troublesome
with regard to these concerted efforts, though, is the mounting difficulty of
distinguishing narcissistic self-promotion from truly engaged activism, and
idealistic from commercial purposes. Activist’s online choreographies seldom
involve just one platform; the goal of most political and ideological causes is to
go viral across online and offline media in order to accumulate massive
attention.39 These short outbursts of massive viral looping may (p.88) illustrate
how the algorithmic syntaxes of liking, friending, trending, following, favoriting,
and other “social verbs” blend into a seamless composition, but that harmonized
algorithmic grammar often masks a cacophony of push-and-pull forces.
And yet it is extremely important to identify the mutually reinforcing
mechanisms and contriving alliances that inform the underlying rationale of
social media’s ecology. Twitter’s partnering with Google is not just an incidental
duet between two microsystems, but signals a system-wide trend of
consolidating and confounding forces—a topic I will return to in the last chapter.
The transformation of Twitter’s corporate ambitions and strategies over time, far
from being unique, is a recurring trope in the short histories of various
competing platforms. The double paradox that evolved in Twitter’s years of
interpretive flexibility is peculiarly mirrored in other platforms’ arrested
development (as we will see in the case of Flickr) or successful transformation
(as in the case of YouTube). The dynamics between hyperconnected influencers
and followers as well as the power of predictive analytics have attracted
attention to Twitter’s potential as a predictor and producer of future social
Page 19 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
Turning the analytical prism to Twitter’s own future, though, it may be
precarious for the company to predict what happens next. Since the meaning of
microblogging has not stabilized yet and the ecosystem of connective media is
still in great flux, predicting the future is like playing the stock market: you can
monitor all elements meticulously and not be able to forecast turbulence, owing
to the volatility of the system. In the waxing and waning culture of connectivity,
Twitter’s fate is dependent on its interoperability with other microsystems and
also on the equilibrium between owners’ ambitions to exploit tweets and users’
motivation to keep tweeting. Twitter has not yet lived up to Jack Dorsey’s
aspiration for the platform to “fade into the background.” People do not (yet)
take Twitter for granted. A platform is not a phone. A tweet is not electricity. The
period of interpretive flexibility is far from over; it may still be a long journey to
attain the stage of stabilization.
(1) . See C. McCarthy, “Twitter Co-founder: We’ll Have Made It When You Shut
Up about Us,” CNet, June 3, 2009. Available at http://news.cnet.com/
8301-13577_3-10256113-36.html. Last checked May 16, 2012.
(2) . For the announcement of Twitter’s member milestone, see Mashable: http://
mashable.com/2012/02/22/twitters-500-million-user/. An update on the latest
Twitter figures, both in the United States and worldwide, see http://
www.quantcast.com/twitter.com#summary. Twitter is ranked number 8 in the
global Alexa rankings. See http://www.alexa.com/siteinfo/twitter.com. Last
checked May 8, 2012.
(3) . In an impressive piece of ethnographic journalism, Joe Hagan interviews
Twitter’s CEO and observes the “management control room.” Managers explain
the platform’s transforming ambition, from a social network into an information
network. See J. Hagan, “Tweet Science,” New York Magazine, October 2, 2011.
Available at http://nymag.com/news/media/twitter-2011-10/. Last checked May
16, 2012.
(4) . Twitter’s capacity to deliver messages to various different hardware
platforms has been essential to its success. In a Pew Internet study, Lenhart and
Fox (2009) reported that Twitter users are most likely to access the service
through wireless Internet on mobile devices.
(5) . Early research examines Twitter’s usage as a tool for daily conversation
between friends and for sharing information and news alerts at a community
level (Java et al. 2007; Mischaud 2007).
(6) . Behavioral scientists Zhao and Rosson (2009: 243) concentrate on Twitter’s
role as an informal communication medium in the workplace; they find that the
web service can be used to enhance a feeling of connectedness and to build
common ground for collaboration. Information scientists Honeycutt and Herring
Page 20 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
(2009: 9), while acknowledging that Twitter may not have been especially
designed for informal collaborative purposes, suggest that “design modifications
could make microblogging platforms such as Twitter more suitable for
(7) . Over the past decade, there have been over 100 microblogging sites active
worldwide. Besides Twitter, sites such as whatyadoin.com, Tumblr, Beeing,
PingGadget, Jaiku, and Plurk emerged and disappeared over the years.
Competing services usually occupy specific niches of microblogging,
incorporating various SNS elements; Plurk, for instance, combines
microblogging with video and picture sharing. Pownce integrates microblogging
and file sharing.
(8) . For an introduction to the new Twitter interface, see, for instance, J. O’Dell,
“Here’s a first Look at the New-New Twitter,” VB News. Available at http://
venturebeat.com/2011/12/08/new-new-twitter/. Last checked May 16, 2012. As
Twitter’s CEO, Jack Dorsey, commented in VB News: “A lot of this is around
education…. We saw the same thing ten years ago when people first
encountered URLs. [People needed] an interface that made entry into some of
these new things easier, less scary…. Our users invented this syntax, and we’re
honoring that.”
(9) . For the list of most notable Twitter uses, see Wikipedia. http://
en.wikipedia.org/wiki/Twitter#cite_note-72. Last checked August 14, 2010. This
link is no longer available.
(10) . Evan Williams, one of Twitter’s cofounders and chief executives, said in an
interview with the New York Times: “Many people use it for professional
purposes—keeping connected with industry contacts and following news….
Because it’s a one-to-many network and most of the content is public, it works
for this better than a social network that’s optimized for friend communication.”
See C. Miller, “Who Is Driving Twitter’s Popularity? Not Teens,” New York Times,
August 25, 2009, technology section. Available at http://www.nytimes.com/
2009/08/26/technology/internet/26twitter.html. Last checked May 16, 2012.
(11) . See A. Lipsman, “What Ashton vs CNN Foretold about the Changing
Demographics of Twitter,” ComScore, September 2, 2009. Available at htt://
blog.comscore.com/2009/09/changing_demographics_of_twitter.html. Last
checked May 16, 2012.
(12) . According to a statistical analysis of some 300,000 Twitter users in 2009,
Harvard researchers Heil and Piskorski found that men comprise a minority of
users (45 percent), while they have 15 percent more “followers” than women
and also have more reciprocated relationships. Both men and women are more
likely to follow men than women; in fact, an average man is 40 percent more
likely to be followed by a man than by a woman, while both tweet at the same
Page 21 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
rate. This gender division is unlike other social network sites, where most of the
activity is “focused around women and where men follow content produced by
women they do and do not know, and women follow content produced by women
they know” (Heil and Piskorski 2009, n.p.).
(13) . According to Joe Hagan, in his article “Tweet Science” (see note 3), a
recent Yahoo study found that in a random Twitter user’s feed, roughly 50
percent of the tweets came from one of just 20,000 users. Twitter’s management
duly recognizes the impact of a small number of so-called “hyperconnected”
twitterers (also known as “power users” or “influencers”) as one of its greatest
(14) . Information scientist Christian Christensen (2011) sharply interrogates
Western governments’ enthusiastic promotion of American social media brands
as a technological boost to democratic struggle, raising fundamental questions
regarding the “increasingly blurred lines among policy, development aid,
technological determinism, and commodification” (250).
(15) . The considerable drift toward popular Twitter lists and their following is
underscored by public rankings of “The Top 100 Twitterholics Based on
Followers.” Available at http://twitterholic.com/. Last checked May 16, 2012.
(16) . See J. Hagan, “Tweet Science,” New York Magazine, October 2, 2011 (see
note 3).
(17) . The start-up Klout, which started in 2011, offers a service for measuring
individual’s influence across all social networks (Facebook, Twitter, Google+,
LinkedIn); a Klout score is a number between 1 and 100, with scores over 50
beginning to indicate a high degree of influence. For more information on
Klout’s measuring philosophy, see http://klout.com/corp/kscore. Last checked
May 16, 2012.
(18) . A study performed by Pear Analytics in August 2009 (Kelly 2009) analyzing
2,000 tweets over a two-week period found that approximately 83 percent of all
tweets involved short conversational, expressive, and promotional statements
that form the heart of “pointless babble.” This outcome was disputed by social
networking researcher danah boyd, who responded to the survey in a blog post,
stating that “pointless babble” is better characterized as “peripheral awareness”
or “social grooming.” See d. boyd, “Twitter: ‘Pointless Babble’ or Peripheral
Awareness + Social Grooming?” Apophenia, 2009. Available at http://
www.zephoria.org/thoughts/archives/2009/08/16/twitter_pointle.html. Last
checked June 12, 2012. In earlier work, boyd and Ellison (2007) argued that the
flow of tweets reflects a form of “networked sociability” aimed at maintaining
intimate relationships with friends, following high-profile users, and connecting
with other people—close and remote (cf. boyd and Ellison 2007).
Page 22 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
(19) . While early observers emphasized the conversational nature of tweets,
later researchers focused more on tweets’ informational content, by looking at
Twitter’s function either as a headline-news distribution system (Kwak et al.
2010) or as a journalistic tool (Emmett 2008; Hermida 2010; Hirst 2011; Murthy
2011). A Pew State of the News Media report published in 2012 shows that only
9 percent of people get their news from social media. See http://
pewresearch.org/pubs/2222/news-media-network-television-cable-audioo-radiodigital-platforms-local-mobile-devices-tablets-smartphones-native-americancommunity-newspapers. Last checked May 16, 2012.
(20) . As one study by information scientists phrased it: “Twitter has shown how
a medium for social networking and microblogging can be used as both a tool for
delivering essential information, i.e., news, as well as a medium for delivering
non-essential information, i.e., personal messages” (Blake et al. 2010: 1260).
(21) . For stats and figures on Twitter use, see http://blog.twitter.com/2011/03/
numbers.html. Last checked May 16, 2012.
(22) . Twitter’s CEO Dick Costello quoted in J. Hagan, “Tweet Science,” in New
York Magazine, October 2, 2011 (see note 3).
(23) . Rumors of Twitter’s impending takeover by Google or Facebook in early
2011 were quickly denied by the site’s owners. See, for instance, Rupert Neate,
“Twitter Denies $10bn Takeover Talks with Google and Facebook,” Telegraph,
February 14, 2011. Available at http://www.telegraph.co.uk/technology/twitter/
Last checked February 22, 2011.
(24) . Alan Rusbridger, editor-in-chief of The Guardian, explains why Twitter has
become an indispensable news source for journalists by listing 15 features. See
A. Rusbridger, “Why Twitter Matters for News Organisations,” Guardian,
November 19, 2010, http://www.guardian.co.uk/media/2010/nov/19/alanrusbridger-twitter. Last checked April 17, 2011.
(25) . In July 2012, Apple reportedly considered taking a stake in Twitter. Apple
has tightly interwoven Twitter features into its software for phones, computers,
and tablets. I will return to these partnerships between social media, software,
and hardware firms in the last chapter
(26) . Since 2006, Twitter has relied primarily on investments from investors like
Fred Wilson, a venture capitalist and principal of Union Square Ventures. In a
2010 round of funding, six investors, including T. Rowe Price (TROW), Insight
Venture Partners in New York, and Spark Capital in Boston, reportedly pumped
$100 million into the company.
Page 23 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
(27) . Journalist Claire Miller (2009) reports about market analysts getting
impatient with Twitter. See C. Miller, “The Obsession with Twitter’s Business
Model,” New York Times, March 26, 2009, Business section. Available at http://
bits.blogs.nytimes.com/2009/03/26/the-twitterverses-obsession-with-twittersbusiness-model/. Last checked May 16, 2012. See also D. L. Smith (2009),
“Twitter’s Business Model: Brilliant or Non-existent?” Harvard Business Review,
October 26. Available at http://bx.businessweek.com/twitter-business-model/
twitters-business-model-brilliant-or-non-existent/1376625150997754291aecc46fcc0cdeefdd60745fe01e68cb9/. Last checked May 16, 2012.
(28) . Besides Tweetdeck, a San Francisco-based start-up called CoTweet
successfully developed services to manage large companies’ Twitter accounts
(e.g., Coca-Cola) by tracking interactions with customers and letting employees
respond. See C. Miller, “Tensions Rise for Twitter and App Developers,” New
York Times, April 11, 2010, Technology section. Available at http://
www.nytimes.com/2010/04/12/technology/12twitter.html. Last checked May 16,
(29) . Promoted Tweets and Promoted Trends work as follows: if you look at the
right side of users’ Twitter feeds, one “promoted trend” is added to the
traditional top ten of most popular tweeted topics. By adding a paid-for eleventh
trend to the list—for instance Disney-Pixar’s most recent movie title—the
sponsor hopes the item will rise up the list. In addition to Promoted Tweets and
Trends, companies have already started to use Twitter to engage viewers in their
broadcast ads; during the 2011 Superbowl, for instance, car maker Audi
promoted a Twitter hashtag in its commercial, inviting viewers to join a
conversation about the meaning of the concept of “progress.”
(30) . M. Evans, “Is Twitter’s Business Model the ‘Firehose’?” Blog post on
Twitterati, April 26, 2011. Available at http://www.twitterrati.com/2011/04/26/istwitters-business-model-the-firehose/. Last checked May 16, 2012.
(31) . Much to the company’s credit, Twitter offers an archive of its terms of
service on the bottom of its website, which makes it easy to track how their
policies have changed over the years. See http://twitter.com/tos. Last checked
November 3, 2011. All further references to Twitter’s terms of service refer to
this website.
(32) . In Western countries, legislators may demand that social network services
hand over their data, in order to help central governments carry out their own
“citizen intelligence” programs. In 2011, Twitter lost a case where it refused a
subpoena to deliver the tweets of an Occupy protestor to the California courts.
In an unexpected move, Twitter took on the court’s decision by delivering an
ingenious argument why it was not legally forced to deliver these data. Many
users praised the company’s defense strategy on behalf of its users. See, for
Page 24 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
instance, Zach Walton, “Twitter Defends User in Court over Occupy Tweets,”
WebProNews May 10, 2012. Available at http://www.webpronews.com/twitterdefends-user-in-court-over-occupy-tweets-2012-05. Last checked May 16, 2012.
(33) . See, for instance, C. Arthur, “Twitter Faces Censorship Backlash,”
Guardian, January 27, 2012. Available at http://www.guardian.co.uk/technology/
2012/jan/27/twitter-faces-censorship-backlash. Last checked May 12, 2012.
(34) . Twitter director Ryan Sarver quoted by D. Taft, “Users lash out at new
Twitter restrictions.” TechWeek Europe, March 14, 2011. Available at http://
www.techweekeurope.co.uk/news/developers-lash-out-against-new-twitterrestrictions-23678. Last checked May 15, 2012.
(35) . This comment comes from a Google groups user by the name of Klondike.
Available at https://groups.google.com/forum/#!msg/twitter-development-talk/
yCzVnHqHIWo/gYUpkfrGXvwJ. Last checked May 16, 2012.
(36) . In parliamentary elections in the Netherlands in September 2012, Twitter
was claimed to be as accurate as most election polls because it “echoed” the real
voice of the people. See “Politieke peiling kun je net zo goed via Twitter doen,”
De Volkskrant, September 9, 2012. Available at http://www.volkskrant.nl/vk/nl/
Politieke-peiling-kun-je-net-zo-goed-via-Twitter-doen.dhtml Last checked
September 12, 2012.
(37) . Charles Duhigg in his book The Power of Habit (2012) insightfully maps
the market of predictive analytics. One of his persuasive examples is a
mathematician employed by Target who designed an algorithm that predicts on
the basis of 20-some variables whether a female customer is pregnant, and lets
interviewers avoid asking this impertinent question. When the father of a 17year-old girl complained about Target’s uncalled-for ads for diapers and baby
stuff targeted at his daughter, he apologized three weeks later, after he found
out about his daughter’s pregnancy—a state that she had carefully hidden from
her parents.
(38) . See D. Carr, “Hashtag Activism and Its Limits,” New York Times, March
25, 2012. Available at http://www.nytimes.com/2012/03/26/business/media/
hashtag-activism-and-its-limits.html?pagewanted=all. Last checked May 16,
(39) . Such was the case in the Kony 2012 affair, when an activist’s video
produced by Invisible Children urging the indictment and incarceration of
Ugandan war criminal Joseph Kony went viral through Twitter, YouTube,
Facebook, Flickr, and Vimeo and was picked up by virtually every offline news
organization in the world.
Page 25 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020
Twitter and the Paradox of Following and Trending
(40) . Part of a growing market for Twitter-based predictive analytics, companies
like Topsy Labs use the tweet streams of millions of people to help predict the
future—from the spread of disease and financial fluctuations to elections and
revolutions. Using the incessant flow of tweets, marketers get a real shot at
engineering trends, popular topics, and the popularity of people. See M. Ingram,
“Can Watching Twitter Trends Help Predict the Future?” Available at http://
Last checked May 16, 2012
Access brought to you by:
Page 26 of 26
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All
Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.
Subscriber: UC – San Diego; date: 03 March 2020

Purchase answer to see full

Why Choose Us

  • 100% non-plagiarized Papers
  • 24/7 /365 Service Available
  • Affordable Prices
  • Any Paper, Urgency, and Subject
  • Will complete your papers in 6 hours
  • On-time Delivery
  • Money-back and Privacy guarantees
  • Unlimited Amendments upon request
  • Satisfaction guarantee

How it Works

  • Click on the “Place Order” tab at the top menu or “Order Now” icon at the bottom and a new page will appear with an order form to be filled.
  • Fill in your paper’s requirements in the "PAPER DETAILS" section.
  • Fill in your paper’s academic level, deadline, and the required number of pages from the drop-down menus.
  • Click “CREATE ACCOUNT & SIGN IN” to enter your registration details and get an account with us for record-keeping and then, click on “PROCEED TO CHECKOUT” at the bottom of the page.
  • From there, the payment sections will show, follow the guided payment process and your order will be available for our writing team to work on it.