São Paulo, Brazil / October 5-8
Hotel Tivoli

Web 2.0 in Brazil

by Dawn Nafus, Rogério de Paula and ken anderson

This paper was published as “Abstract 2.0: If We Are All Shouting, Is there Anyone Left to Listen?” in the Ethnographic Praxis in Industry Conference Proceedings (2007:66–77).

Abstract: This paper explores notions of ‘voice’ as it relates to Web 2.0. We begin by tracing the social meanings of Web 2.0 technologies in Brazil. There the notions of ‘voice’ as conceived of in the American media are absent, yet significant collective action took place online through a kind of speaking out. Next the paper describes the conflation of voice with a notion of social networks to explain how the American media misread the Brazilian action. This is achieved by an incredible plasticity and abstraction of the ‘Web 2.0’ construct, which flattens otherwise qualitatively meaningful distinctions. This puts us on some ground to raise the issue of how abstractions might become relationships. This, we argue, is evidenced both in terms of how Brazilians might interpret online relationships, and how Web 2.0 hype betrays a politics of abstraction at work in the wider economy.


This paper examines the notion of ‘voice’ in the paradigm of “Web 2.0.” We are interested in voice as it is constituted in the use of such systems, but more so in the public discourse about it. Here we trace the social life of Web 2.0 as a global assemblage (Ong & Collier, 2005), that is, a phenomenon that happens in particular places and times but is not a phenomenon ‘of’ any particular site. We view voice as an ethnographic phenomenon rather than as a theoretical object. That is, we are concerned with who thinks they have voice, and what they imagine it to be, and are not so concerned with creating a theory of voice as it relates to anthropological writing or industrial projects. There is an extended body of literature on the topic and little cause to rehash it here. The discourse about Web 2.0 centers on user empowerment, or the ability of users to make their own content and linkages. In this paper we explore the possibility that the more ‘power’ given back to users, the more that ‘voice’ might be emptied of texture and social significance. Our main argument is that the notion of voice is becoming flattened and divested of power through the ‘brand’ of Web 2.0. Voice is not emergent from what the software ‘does’ or how people ‘use’ it (although these things are nonetheless caught up), but instead is a discursive device that conflates a number of social relations into a reified commodity.

As authors it would be nice to be able to offer a definition of what Web 2.0 actually is. But this would do some violence to it as an emic concept. One can say it is a metaphor people use to describe what is currently normative about computing. There is a new look and feel to the Internet that many attribute to it myriad new possibilities. But beyond this it escapes most moorings. Instead it floats as its own context. Technology has historically been a privileged site of self-induced ‘meaningfulness’ that cares little for the specificity of social relations (Woolgar et al 2000). If it is deemed a technology, these authors argue, it is deemed to have ‘implications’ without having to know what these actually are. So, one should not expect too much in the way of substantive meaning from these computing ‘paradigms.’ Most often, business people point to it as a means of justifying business plans based on hoped-for network effects (because Web 2.0 is about social connection), or attributing agency to users of it (because Web 2.0 is about self-production of media). The term gets used to provide context without specifying a place or people or objects.

We begin by tracing the social meanings of Web 2.0 technologies in Brazil. There the notions of ‘voice’ as conceived of in the American media are absent, yet significant collective action took place online through a kind of speaking out. Next the paper describes the conflation of voice with a notion of social networks to explain how the American media misread the Brazilian action. This is achieved by an incredible plasticity and abstraction of the ‘Web 2.0’ construct, which flattens otherwise qualitatively meaningful distinctions. This puts us on some ground to raise the issue of how abstractions might become relationships. This, we argue, is evidenced both in terms of how Brazilians might interpret online relationships, and how Web 2.0 hype betrays a politics of abstraction at work in the wider economy.

Growth of Web 2.0 in Brazil

Networking as a Collective Experience

In an advertising driven economy, the expansion and retention of eyeballs is everything. It makes a certain sense, then, that narratives of success in Web 2.0 involve tales of technologies skyrocketing from local to global scales. Craigslist, for example attempted to solve a small-scale, local (even personal) ‘need’ as a relative newcomer in San Francisco to connect to other people, and reached unexpectedly large proportions. Orkut started from a US company (Google) to compete with Friendster. It was not local at all but quickly became localized, prompting talk of a “Crazy Brazilian Invasion”, which has been captured and extensively narrated by the Brazilian media. 1 A biographical account of this process can be summed up in three major epochs: ‘Orkutmania,’ ‘Orkuticídio’ (Orkut suicide – as some users realized the impact of turning themselves “public” on their everyday, external lives), and Orkut legal challenges (as, for example, illicit, illegal external practices, such as drug dealing found their way to this ‘virtual’ network of potential users). In January 2004, Orkut in Brazil was no more than a short note on a local newspaper about a new US-based relationship site by Google to compete with Friendster.com. Based on ‘invitation only’ subscription model where new users were required to receive an invitation from a member to join the site, Orkut grew organically out of individuals’ social network paths. These initial users had an interest in Orkut as a social networking system. That is, the number of communities about social network analysis and related topics as well as the types of messages posted on them were relatively high.

While it was growing steadily in the US and starting to reach out overseas, something unexpected happened – a Brazilian tidal wave engulfed the site. 2 Five months after this first newspaper note about this, the number of Brazilian users ‘beat’ the US user population on site. This would have gone unnoticed if were not for that the ways in which Brazilians challenged pre-established social norms (or netiquette) and technological assumptions embedded in the design. On this note Brazilians took it as a clean slate, so to speak – although the system’s UI was designed to afford the particular notion of social network that pervaded US academics and businesses at that point, Brazilians had not necessarily a pre-established set of norms and behaviors in place to shape and organize the use of the system. Instrumental notions of social network management described in the sociological literature (Burt, 1995; Granovetter, 1973; Nardi, et al., 2002) as critical aspect of people’s everyday work practices turned out to be uninteresting and even unimportant for Brazilians adopting Orkut. In fact, people at that point had not been exposed to the existing instrumentalist discourses of social-network and social networking common in the US – the notion that a person fans out to collect as many contacts, and therefore opportunities, as possible, and that an affective relationship might be a totally separate concern. Orkut was not a social networking system but a site for “networks of relationships” (direct translation of “redes de relacionamento”) or simply a relationship site. Although Orkut visualized relationships in this mathematical way, reflecting its instrumentalist origins, this did not stop people from viewing these as relationships rather than a disembodied network.

The Brazilian “takeover” in fact generated a big (negative, for the most part) reaction from other users that ranged from xenophobic hate messages to puzzled concerns. According to (Reuters, 2004), a Canadian user, Tammy Soldaat, was quickly labeled ‘nazi’ and ‘xenophobic’ by a group of Brazilians for asking whether only people who speak English should participate on her community site on body piercing. Brazilians did not hesitate to push off people (usually non-Portuguese speakers) from communities. Yet they argued that they simply wanted to “hang out” with their friends, as they usually do in bars, on the street, or on the beach. “Since we can invite anyone we want at Orkut, and my friends are Brazilians, it doesn’t make sense talking to them in English,” Reis said in Portuguese, “I use the language I know” (Quote from Reuters, 2004). Hence, in what terms were Brazilians acting (or behaving) differently than the rest of site user population?! The increasing frequency in which users start seeing postings in Portuguese told them that something unusual and unique was taking place – the realization that Americans are no longer necessarily the predominant (or dominating) user population on the web.

Becoming the largest geographic community, gave Brazilians a sense of ownership. This ownership was without any legal stake or control. There was a perceived right to overlook some of the basic community netiquette, such as, not respecting a community’s language, as well as to complain about changes on the site UI, configuration, login setting, etc., which Brazilian Orkut users did amply. On the other hand, as one American user complained “Orkut is not even a Brazilian service” – so where do all these ‘rights’ come from? While Google had (and still has) a business interest in keeping Orkut running, nothing (legally or otherwise) could prevent them from discontinuing the service, changing policies, changing the technology, restricting access, or even charging for its uses.

Today, it is estimated that Orkut has reached over 63 million users, where 68.3% of its access comes from Brazil, according to Alexia, 3 14.6% from India (the fastest growing user population on Orkut, today), and 3.5% from the US. These numbers are even more significant when compared to participation of these countries on the Internet as a percentage of the worldwide Internet user population: the US leading with 18.8%, India coming in 5th place with 3.6%, and Brazil in 8th with 2.9%. By far, Orkut is the most visited site in Brazil, second most visited site in India, and merely 31st place in the US.

We can think about the nature of this takeover by contrasting it with two common views of social networking software: the individual and the collaborative. The individual view focuses on actions and experiences of solidarity actors and independent decision-makers. Here the focus has been on instrumental uses of social network software. This view would surface stories about how individuals from urban areas around Brazil were using Orkut to connect to past classmates to supplement their job opportunities or make new friends in places they wanted to visit, as well as the usual making connections on regular basis. This all was indeed taking place. The collaborative view of social networking views it in terms of orchestration and making something happen in collaboration with others. Using Orkut to assist in arranging meeting times for evenings or weekend get-aways with groups of friends might be an example. Being connected to others interested in sports for the physically impaired is another. Again, we have traced these activities through our fieldwork. An issue with collaboration here is that the social aspect still assumes a multiplicity of individuals, without looking at the collective as a whole. What the ‘takeover’ of Orkut offers a case study collectivity rather than collaboration. By collective experience we are referring to the intersubjectively negotiated, individually incorporated, and only more or less “shared”, and yet in a way forms a common lens through which we make sense of an everyday experience.. Orkut grew in this way to be a part of the everyday experience of all the middle class participants in our studies between 2004 and 2006. The use of Portuguese as a voice to gain power vis a vis the Americans was the exciting aspect of it.

Orkut and Emergent Forms of Social Relations

As the takeover was happening, a set of transformations was taking place. The meaning of various social categories, ranging from kinship categories, to hanging out with friends, to private, personal conversation were shaped by the ways in which the system represented such concepts in its technological affordances (e.g., UI), and consequently people endowed new meanings to these patterns of interactions mediated by these affordances. For example, conversations were re-framed as ‘posting’ messages (personal or otherwise) on people’s (acquainted or otherwise) public ‘scrapbook,’ (scrapbooks are a form of offline chat or messaging system that are one of the most popular activities on Orkut.) Writing public messages on scrapbooks parallels how people interact and approach other in public spaces in the physical world. Friends, high school buddies, relatives, co-workers, acquaintances, and the like, became equally unmarked as buddies in people’s buddy lists – or social network. This is a vast difference than what happens in physical world where categories amongst out participants in research were multitudinous. Distinctions that were often situational and dynamic in real life, in Orkut were static and fixed. The result was the categories became relatively meaningless. What became important was adding people to the buddy list. This was not a marker of one’s personal set of relationships but rather one’s “buddy list” became an entity in and of itself – the value was to have a large buddy list and the set of relationships was mostly irrelevant. While in the US users had a restrictive, exclusive notion of a personal social network – i.e. personal circles – users in Brazil had an exceptionally open, inclusive notion, such that it did not matter who was ‘in’ and who was ‘out’, what was ‘publicly’ available and what was not. What did matter was the size of the buddy list. It became a new social artifact.

While the distinctions of ‘friend,’ ‘kin,’ ‘fellow party goer,’ etc were not challenged in everyday life, in online life the notion of a ‘buddy list’ began to take on its own meaning. For the first time, the quantity of contacts, in such starkly reified form, became sociologically meaningful. In fact, users started somewhat competing for the number of people in their buddy list. For example, Brazilians on average had 150 other members in their buddy list whereas in the US people had around 40 or 50 buddies. One of our interviewees, while showing her buddy list, was unable to recognize people in her own list. That these buddies are listed and visually present construes a different way of thinking and interacting with people. At a party, for example, value does not come necessarily from the counting of the number of people but how enjoyable it is, as they put in Brazil, “festa animada.” The shift here was to an accounting where counting actually mattered in the virtual environment of Orkut. Similarly, Burrell (2007) notes how video technology in Ghana is in fact used not just to send well-wishes across a diaspora, but to count, and account for, who was present at gatherings. Orkut became a way to count and to account for who is present among your social relations. Our argument is not that accountancy is anathema to sociality in a general way. The competitive gifting of yams in Papua New Guinea, stacked and lined up in dazzling displays, famously demonstrates otherwise. However, we do claim that information technologies are particularly good at accounting, and such social accountancy has not historically been a part of Brazilian sociability. On Orkut, counting took on a social resonance. A person as a central node in the network was both more visible and more important. We can think of this transformation as a kind of audit culture (Strathern 2000). Audit practices seek to measure production, but the fact of measuring changes the dynamic of what is measured. Education becomes a matter of exam passing and time spend doing degrees; teachers then focus on passing standardized questions and shuffling through students rather than producing critically thinking persons. Similarly, visualizing social networks into lists and nodes and spokes is not a mere mapping of what already exists, but the very act of measuring has consequences for social relations.

Numbers became a part of not just individual connectedness, but politics of what constituted a ‘community.’ According to recent market research figures just 9.8% of Brazilian households are connected to the Internet (Jupiter Research, 2006). The number of registered user far exceeds this figure. Orkut has become so pervasive that in Brazil when you ask about who is connected people will respond ‘everyone.’ In practice, Orkut was being accessed by people who were not normally consider key players in the informational revolution, secretaries from work, dental hygenists at neighboring offices, girlfriends at friend’s homes, and even women moving into the formerly male dominated sphere of public LAN houses. Orkut has been used in Brazilian LAN houses as a way of including (otherwise bored) girlfriends in boy’s computer activities. Speaking to a former LAN house owner, he told us that he used to send out Orkut invitations to the girls hanging out with their boyfriends in his LAN house, as a way to encourage them to stay longer. Initially, only boys would hang-out there, for the most part, playing games, but they would not stay long because their girlfriends would quickly get bored waiting. Soon, he realized that in giving access to Orkut, girls would spend their time browsing, chatting, checking other people’s pictures on the site, letting their boyfriends play longer ‘in peace.’ In the end, he reported more girls coming to his LAN house to access Orkut than guys playing games. The on the ground practice of Orkut was so pervasive by 2006 that if you were not on Orkut, you are out – you are left out of external social life. Many are the accounts of people who were not invited to a friend’s birthday party because they were simply not on Orkut or in a person’s buddy list. As such, it became its own form of social connection over time.

With such growth, it became increasingly normative to be on the site. The Orkut invitation-only subscription model endowed the site as a whole with notions of being part of ‘the’ network (or community). A certain type of social status created an aura around the site – “just wait, because sooner than later some one in your social network will send you an invitation.” Without any question, this drew great attention and excitement to the site, and people expressed anxiety getting access to it. At one point, people started selling invitation on a local online auction site (prices ranging from a dollar to US$15). What really excited people, however, was competition for ousting the US as the largest user community. It started as a game (probably as an imagined soccer game of sorts against the US), where a Brazilian user created a community challenging the Brazilian user community to invite as many other Brazilians as possible until they achieved “the first place” in number of users according to a particular geographical origin. Orkut used to have a demographics webpage where users were able to see a number of very simply statistics about the user population according to self-defined personal information (e.g., geographic location, language, age, gender, and the like). The highlight of this page was the ranking of countries according to their number of users—next to the country’s name one finds its flags, reinforcing the soccer game imaginary. On June 23rd 2004—a memorable date, according to the broadcast news (yes, it had national media coverage4)—Brazilians (31.55%) took the leadership of Orkut, surpassing the Americans (29.62%)—at that point, the site had not reached the 1 Million mark.

‘Community’ was used to describe online circles of people, sometimes in the sense of community of interest, as originally intended, but increasingly something more pliable. While there was an incredibly majoritarian and inclusive notion of community at work, it also became more suffused with numbers than the stuff of social relations. The numbers were the politics, rather than questions of, say, leadership or group provisioning or the myriad other things that people do in social groups. Community, of course, has always been a slippery word that people have to reinvent for themselves constantly. But here people were not creating its meaning out of relations, but of ever increasing numbers on a statistical reporting page. Again, this is not to say that social relations did not take place, or that Orkut was in some way asocial, but that the display of numbers grew its own social life such that it became interchangeable with notions of what a social circle was and what an interest group might be. With so many people on buddy lists and joining ‘communities’ within Orkut left and right, it also was interchangeable with a notion of a personal network—now equally arbitrary. As Brazilians were gaining a voice on Orkut, the very thing that gave them that power—a percentage—also eclipsed any substantive meaning of those links.

Voice and the Plasticity of Web 2.0 as a Brand

We became interested in Orkut because it is a great success that remains utterly obscure to Anglophone discourse. It may as well be a passing comment or a footnote in the literature about social networks. The story about Orkut in the US should have been about the success of a global technology becoming local, but there was barely any story at all. What mention there is treats it as a negative example around issues of ‘misuses’ and focuses its legal troubles with local authorities. In this section we look at this absence in light of what was being said about Web 2.0 in the Anglophone press. One of the hallmarks of Web 2.0 discourse is the notion that seemingly anyone can have a voice on the web. “You” are the person of the year, according to Time Magazine, because “You” made videos, avatars, filled in Facebook content and design competitions. By doing this “You” used “a tool for bringing together the small contributions of millions of people and making them matter” (Time, 2006). Here social connection was assumed to be the site of knowledge production. Homemade videos are interesting if others link to them; websites like SecondLife allow you to mash up other’s digital parts to make an avatar. As mainstream media, Time perhaps had a particular reason to be concerned about how we ‘needed’ this revolution to get beyond pre-digested newsbytes. The new technologies they said made information more direct. Similarly, the cover of Newsweek some six months before declared Silicon Valley was “Putting the We in Web”, again under the same rubric of consumer empowerment, circumventing broadcast media. Together the news reports and blogosphere chatter hailed in a new era of openness.

What made Orkut impossible for an American public to even pay attention to was that in the Orkut case, such voice does not come via an individual ‘expressing themselves’ through their online generated content. Far from it—most Brazilian Orkut users did not have the faintest imagination of the publicness of their writings. They were not lone voices shouting out to the world. The Anglophone press betrayed a commitment to at most a collaborative, but certainly not collective, view of social networking. What it was particularly concerned with was the individual as an expressive agent. The stories assumed each atomized person made their own contribution. The mechanics of “making them matter” are consistently left wooly in these reports. A loose, atomized version of connectedness in the various software systems themselves provided an imagined stand in for counting socially.

Brazilian ‘voice’, however, was heard in the various social network norm violations and competitive recruitment until it did become dominant. The being heard was the decolinalization of the Web. It was a demarginalization of a non-English language as a dominate language of the Web and the emerging of the collective power of an emerging society. This was decidedly NOT the sort of voice imagined when Time Magazine nominated ‘You’ the Person of the Year.

There are enormous cracks in the just-so stories presented by Time and others. The most frequently watched videos on YouTube, for instance, are just as likely to be clips from broadcast programming and other commercially produced footage. Blogs are put up at astounding rates, as are MySpace pages, and they go dead just as easily. Network television viewership is declining, but it is somewhat of a leap to suggest a new participatory media is at stake, as if there were no participation in mass media previously, or ways in which people made it relevant to themselves. YouTube and the like are construed as ‘participatory’ because a slippery vocabulary of ‘community’ has emerged which places an ever expanding set of behaviors and phenomena under the same umbrella as if it were the same thing. Somehow, the Web 2.0 paradigm excavates democratic expression out of the minutest communication act.

Pages: 1 2