Linking Internet Texts and Practices: Challenges and Opportunities of Interdisciplinarity in an Ethnographically Inspired Study of "Local Content"

Can ʻ ʼ zines, some forms of hypertext, continue to be understood as At what point do these ʻ texts ʼ become something entirely different? If so, what do they become, and what might the phenomena in question mean to the ethnographer, the literary theorist, and their respective disciplinary traditions? (Eichhorn, 2001, 576)


Introduction
Ethnographic and ethnographically inspired approaches are becoming increasingly popular in studies of digital media and digital culture, and are being used by scholars from a variety of disciplinary backgrounds, as well as in interdisciplinary projects. Nonetheless, specific methodological (and ethical) tensions and dilemmas can arise in the encounter between different research traditions. One area of such tension relates to how texts are approached, and how they are linked to other types of data. Research into digital culture often involves an encounter with digital texts of one kind or another, reflecting the ongoing centrality of writing in contemporary life as well as the emergence of new writing practices associated with the spread of digital technologies (Barton & Papen, 2010, 3). However, within anthropology, the discipline in which ethnography originates, there has been a tendency for texts to be rather neglected as a form of data (Barber, 2007;Blommaert, 2008;Bird, 2010). Within the 'interdiscipline' (Silver, 2006, 4) of internet studies, there is also a relative absence of methodological and ethical approaches specifically relevant to the study of texts, in contrast with those oriented to human subjects (Bassett & OʼRiordan, 2002, 239). Equally, approaches originating in literary and cultural studies seem not to be equipped to deal with data on (internet) practices, collected through interaction and observation.
This article reports on related methodological questions which arose in interdisciplinary research into how residents of a favela, or shantytown, in Rio de Janeiro, Brazil used the internet, and particularly blogs, to put forward their own representations of the area where they lived. i In recent years, Brazilian favelas -and some of their residents -have gained increased cultural visibility in Brazil and beyond. Music, literature, photography, journalism, videos and other works produced by favela residents themselves -and more broadly speaking by residents of urban periphery neighbourhoods in Brazil -have multiplied, alongside mainstream productions depicting favelas.
The production, dissemination and sometimes wide circulation of work originating in the urban periphery has much to do with increased access to digital technologies, including the internet, in Brazil. The country has a dynamic and innovative digital culture, and access to the internet in its urban areas almost doubled from 2005 to 2010, rising from 24% to 45% of the urban population (CETIC.br, 2006;2011a). Brazilian internet users are also known for their intensive and innovative use of the growing range of interactive platforms available to support the production of usergenerated content on the internet (Fragoso, 2006;Horst, 2011). While overall figures for internet use in Brazil continue to mirror socioeconomic and regional inequalities, the recent increases in access to the internet specifically by favela residents, ii have allowed for the emergence and dissemination of more diverse user-generated content about favelas on blogs, social network sites, microblogs and other platforms.
In the research under discussion here, such user-generated content was understood as ʻlocal contentʼ, or ʻthe expression of the locally owned and adapted knowledge of a community -where the community is defined by its location, culture, language, or area of interest' (Ballantyne, 2002, 2). In some cases, local content about favelas on the internet has emerged thanks to the support of non-governmental organisations and social movements which have established web projects to offer more realistic and diverse representations of these spaces, in response to biases in the mainstream media. However, much of this local content is increasingly also the outcome of personal and collective initiatives, developed more independently. The research focused on this latter type of production, and looked particularly at texts and images, and occasionally videos, published and disseminated on the internet by favela residents, which explicitly engaged with the representation of place. Whilst the study centred its analysis on content published on blogs, this was undertaken with the awareness that dissemination and interaction around blogs often involves other platforms and media including social network sites, microblogs and email, as well as sometimes print formats and face-to-face communication. This has been helpfully termed the ʻblog-circuitʼ by Adriana Braga (2008, 47), iii but other researchers have also recognised the mobility of internet usersʼ practices (and thus content) between and across different platforms (Baym, 2007, n.p.;Postill & Pink, 2012, 10) The aim, then, in this ethnographically inspired project situated broadly at the intersection of Brazilian/Latin American studies and internet studies, was to gain a contextual understanding of the ʻmedia texts of ordinary citizensʼ, which are so often overlooked in scholarship (Rodriguez, 2001, 4). Specifically, the research sought to explore how residents of favelas were representing their neighbourhoods in public internet content, and the significance of these user-generated representations for understandings of the city of Rio de Janeiro. The approach taken went beyond textual and visual analysis of internet content, and complemented this with data on the perspectives of content creators collected through direct interaction and interviews, as well as observational data about internet practices. Although the study focused on the creation and publication of content, and did not explicitly include attention to the reception of that content, it nonetheless developed an awareness of the circulation of the narratives through its dual focus on texts and practices.
This dual focus required methodological and ethical reflexivity, but adapted methodological concepts made it possible to establish creative and constructive connections between the different methods and types of data. To illustrate these connections, this article discusses the conceptualisation of the research process as one of ʻfollowing the contentʼ, inspired by George Marcusʼs (1995) ʻmulti-sited ethnographyʼ, which also approached specific clusters of content on the same topic as ʻcontent eventsʼ. The latter term draws on the idea of ʻliteracy eventsʼ, associated with new literacy studies (Barton & Hamilton, 1998). Overall, incorporating and adapting approaches from different fields made it possible to go beyond an imagined divide between research into texts (within literary and cultural studies) and research into practices (within the social sciences), as well as the choice this implied between ʻfollowing the contentʼ and ʻfollowing the peopleʼ.
Interdisciplinarity was thus both the source and the solution for some of the issues that arose in the course of the research. However, ethnography remained a key influence on the research, and the article thus begins with an exploration of contemporary understandings of ethnography. It then explores the trope of following as a methodological practice within ethnography, suggesting this is highly suited to research in digital culture, before turning to an examination of the status of texts in anthropology and internet studies, and a brief discussion of the increased focus on practices in some areas of media anthropology. The final section draws on work in new literacy studies, which also offered a way forward in connecting texts and practices. I conclude by arguing that the interdisciplinary nature of my research provided an opportunity to adapt and develop methodological concepts from different traditions, in response to the characteristics of my field site.
These framings of methodological practice were both conceptual and practical in nature, enabling me to link different methods and data, and to gain an in-depth, contextual understanding of local content produced by favela residents.

An ethnographically inspired approach
My approach could not be called ethnographic in the traditional sense. I therefore join other nonanthropologists researching the internet using a variant of ethnography (see for example Hine, 2000;Franklin, 2004;Androutsopoulos, 2008) in crediting the ethnographic inspiration of my approach, at the same time as I present caveats about its limitations. Overall, the design and implementation of the study were strongly influenced by scholars using ethnographic approaches to study topics such as the internet (Hine, 2000;Miller & Slater, 2000;Burrell, 2009), grassroots texts of different kinds (Eichhorn, 2001;Blommaert, 2008) and literacy, or writing (Barton & Hamilton, 1998;Barton & Papen, 2010). It also sought to follow certain guiding principles common to ethnographic approaches, regardless of the object of study, namely the need for methodological adaptiveness and reflexivity, and ʻa commitment to try and view the object of enquiry through attempting some kind of alignment with the perspective of those who participate in the researchʼ (Horst & Miller, 2006, 167). However, the research largely did not involve traditional long-term immersion and participant observation, considered by some scholars to be one of the fundamental tenets of ethnography (Miller & Slater, 2000, 21-22;Hammersley & Atkinson, 2007, 3;Estalella & Ardèvol, 2007, 4), and one which is subject to ongoing debate in media ethnography (Murphy & Kraidy, 2003;Murphy, 2011).
At the same time, as this article discusses, renewed understandings of the nature of ethnography, and thus of the ethnographic field site, have been proposed in recent years as the focus of anthropological research has shifted increasingly towards more familiar, urban and mediated settings and topics (including the internet and social media), and as scholars from other disciplines have increasingly adopted the approach in their work (Marcus, 1995;Hine, 2000;Amit, 2000;Hammersley & Atkinson, 2007;Burrell, 2009;Postill & Pink, 2012). As Christine Hine has observed, such developments ʻopen up a space for thinking about ethnography as an experientially based way of knowing that does not aspire to produce a holistic study of a bounded cultureʼ (2000,10). She also notes that ʻethnography of, in and through the Internet can be conceived of as an adaptive and wholeheartedly partial approach which draws on connection rather than location in defining its objectʼ (Ibid; emphasis added). Similarly, Jenna Burrell (2009, 185) has noted that one broader methodological contribution of internet ethnographies has been to demonstrate ʻthe possibility of awareness and analysis of spaces beyond what can be physically inhabitedʼ in fieldwork.
These reframings of ethnography put forward by Hine and Burrell, and others, reinforce the idea that improvisation and responsiveness to the conditions and characteristics of the field can be considered a defining feature of ethnography, over and above any specific method in itself (Hine, 2000;Leander & McKim, 2003). As linguistic anthropologist Jan Blommaert put it, ʻ[w]hen it comes to methods, ethnography has always been characterised by eclecticism and bricolage: the ethnographer thinks and develops methods in response to the features of the object of inquiryʼ (2008,13). In fact, Blommaert uses this assertion to argue that his own study of two grassroots texts from the Democratic Republic of the Congo can be considered ethnographic even though it did not involve fieldwork per se. Kate Eichhorn (2001), whose own research looked at the ʻtextual communityʼ of ʼzines, also questions the inevitability of travel in the ethnographic research experience. Whilst her project involved fieldwork, the bulk of this was conducted from her home.
Eichhorn argues that her research nonetheless remained faithful to a key ethnographic principle in mirroring the practices of the people she was researching, as she requested ʼzines by post and corresponded with ʼzine producers.
I also found that my home (in Brazil) -and my laptop computer -was an important location for my fieldwork. It was the main place from which I followed and ʻlistenedʼ (Crawford, 2009) to content, as I explain later in this article. Indeed, an ʻexperiential rather than physical displacementʼ (Hine, 2000, 45) has become a common feature of contemporary ethnographic research into the internet, although different kinds of face-to-face encounters and events may also persist as important components of fieldwork, depending on research questions and objects of study.

Multi-sited ethnographies and the trope of ʻfollowingʼ
The idea of ʻfollowingʼ as a methodological practice forms part of the ʻmulti-sitedʼ or ʻmobileʼ ethnography proposed by Marcus (1995), which has proved particularly influential. As Marcus explains, in an oft-quoted passage, multi-sited ethnography is designed around chains, paths, threads, conjunctions, or juxtapositions of locations in which the ethnographer establishes some form of literal, physical presence, with an explicit, posited logic of association or connection among sites that in fact defines the argument of ethnography. (Ibid,105) In this understanding, the field becomes dispersed and is more self-consciously constructed by the ethnographer (Amit, 2000). Multi-sited ethnography also allows for different possible approaches to defining oneʼs object of study namely: following the people; following the thing; following the metaphor; following the plot, story or allegory; following the life or biography; following the conflict (Marcus, 1995, 106-110). In addition, Marcus (Ibid,[110][111][112] mentions the possibility of undertaking a ʻstrategically situated (single-site) ethnographyʼ, which may involve less movement, but still includes an explicit awareness of, and attention to, multi-sitedness.
The latter formulation was one of the key influences on Burrellʼs study of the appropriation of the internet in Accra, Ghana, and her experience in turn led her to put forward the concept of ʻthe field site as networkʼ (2009,195). In this configuration, the field site becomes defined by the ʻphysical movements, places indexed in speech and text, and social imaginings produced by research participantsʼ (Ibid,196), which the researcher both follows and intercepts. The field site is thus ʻa network composed of fixed and moving points including spaces, people and objectsʼ (Ibid, 189), although not all of the spaces are physically accessible to the researcher (or indeed to research participants). There are echoes here of Sarah Straussʼs understanding of the ethnographic field, based on her study of transnational yoga practices, ʻas an intersection between people, practices and shifting terrains, both physical and virtualʼ (2000, 171-2).
The trope of ʻfollowingʼ is thus central to the work of both Strauss and Burrell, although Burrell stresses that following is not always a physical act, and can take place ʻthrough space as well as in languageʼ (2009,196). Straussʼs yoga study involved her in ʻfollowing threads and trails of people, publications and practices that together told a storyʼ (2000,161). Kevin Leander and Kelly McKim also reflect on ethnography as a mobile practice, and bring together influences from literacy studies, spatial theory and ethnography in sketching out how researchers could ʻfollow and interpret space-constituting practicesʼ (2003,225), including through ʻthe mapping of flow, such as the flow of texts and bodiesʼ (Ibid, 225). Like the researchers cited above, and like Hine (2007), I thus sought to be alert to connections and predisposed to mobility (whether methodological or analytical, physical or digital) in my own study, which ultimately focused in on the work of just three content creators from a single Rio favela.
Whilst my research remained centrally oriented to place, given my interest in ʻlocal contentʼ and the digital representation of place, siting my fieldwork in the favela where the content creators I was following lived represented only one available strategy among many. In the early stages of the research, I mapped internet content being produced by residents (and to a lesser extent, projects or institutions) in the area, searching for and documenting blogs and websites, as well as social network site communities (groups) with a geographical or thematic orientation to the favela. I also undertook exploratory observations in small internet cafés in the area (known as lan houses in Brazil). This exploration of different potential entry points helped in narrowing down my focus and ultimately in constructing a more viable and focused field site. Having clarified and confirmed that my interest lay in content which explicitly offered representations of a specific favela from a resident's point of view, produced at least potentially for an external audience, I chose to site my field principally in this type of content, which I had found primarily on blogs. I thus undertook ʻsystematic observationʼ (Androutsopoulos, 2008) of blog content, and followed the flows and connections which presented themselves through that content, including opportunities for direct encounters with its creators, whether face-to-face or via other channels (email, internet chat).
Despite the focus on blogs, however, it was messages posted in a forum on the social network site Orkut, highly popular in Brazil at the time, which played a crucial role in leading me to blog content which later featured in two of my case studies, content which in turn led me on to other related material. For me, like the researchers mentioned above, it therefore worked well to conceive my fieldwork as a process of following, in this case following the travelling texts and practices of local content creators, as these moved across internet platforms, across media, and across the city itself. The trope of following thus became ʻfollowing the contentʼ in my approach.
In fact, incorporating ʻfollowingʼ into the methods used in a study of the publication and dissemination of internet content made methodological sense not only due to the relevance of Marcusʼs multi-sited ethnography (and those who have adapted and developed it further), but also because following itself has come to the fore as a mode of engagement in contemporary internet platforms and practices. It is particularly associated with the micro-blogging site Twitter, in which users opt to ʻfollowʼ and therefore automatically receive each otherʼs postings, but is also commonly used in relation to slightly older technologies such as RSS (Really Simple Syndication), which allow users to be automatically informed of updates to the content of selected websites, without having to manually check sites on a regular basis.
Viewing my fieldwork as a process of ʻfollowing the contentʼ thus enabled me to keep track not only of the appearance of new posts on particular blogs, but also of the evolution of the design and layout of blogs, including what I call ʻframing contentʼ (the visual and textual elements of blog sidebars, headers and footers, as well as static blog pages, surrounding the flow of dynamic blog posts), which also changed frequently. This attention to the detail of content was particularly important given the mutability and openendedness of blogs as sites of publication and interaction (Himmer, 2004;Moody, 2008;Reed, 2008). For this reason, it was essential also to view blog posts in situ (rather than only in my RSS reader, where they appeared detached from their original publication site), as well as to take regular screenshots, which later enabled me to engage in detailed analysis of how content evolved. A further dimension of the ʻfollowing the contentʼ approach was an attempt to trace the dissemination and publication of content across different internet platforms as well as in print, following the links embedded or implied within content and the paths which content took beyond its original site of publication through its dissemination and reposting.
Rather than seeking to participate in internet content creation myself, as some researchers have done (Estalella & Ardèvol, 2007;Beaulieu & Estalella, 2012), my mode of engagement was closer to what Kate Crawford (2009) has termed ʻlisteningʼ, in her reframing of ʻlurkingʼ -a more pejorative term usually associated with passive observation -in the context of social media. Whilst she does not write about listening specifically in terms of the fieldwork roles available to social media researchers, Crawfordʼs observation that listening ʻinvokes the more dynamic process of online attention, and suggests that it is an embedded part of networked engagement -a necessary corollary to having a "voice"ʼ (Ibid, 527), fits well with the way I sought to follow and be receptive to the work of local content creators.
Even with the trope of following in mind as a key principle, however, determining what or who to follow proved a challenge. In the early stages of my project, the question was whether to begin with the local content about a particular favela which could be discovered and accessed via the internet (ʻfollowing the contentʼ), or whether to begin with a specific physical location in the favela (such as an internet café), and explore the content being generated by its users (ʻfollowing the peopleʼ). I concluded that the approaches could be fruitfully combined, and were likely to overlap naturally, but ultimately gave prominence to the ʻfollowing the contentʼ formulation. In practice, it became clear that following content necessarily involved me, to a certain extent at least, in following the creators of that content, due to my interest in the practices of content creation. This uncertainty about the distinction between following content or people thus relates directly to the methodological challenge discussed in this article, namely the attempt to link internet texts and practices.
Another key issue in my study related to the analytical stance to be taken towards internet content.
The internet is often understood by researchers as a ʻspaceʼ or ʻsiteʼ for interaction, and yet, as Elizabeth Bassett and Kate OʼRiordan have pointed out, it is also ʻa medium through which a wide variety of statements are producedʼ (2002,234). However, dealing with internet texts in interdisciplinary research informed at least in part by social sciences approaches, where there is also engagement with the creators of those texts, can be challenging given the relative lack of established methodological and ethical approaches in this area. Such an intention confronts headon the ʻtensions between social texts and social spaces, and between representations and peopleʼ (OʼRiordan, 2010, n.p.) in internet research methodologies. Nonetheless, Bassett and OʼRiordan (2002, 244) conclude that both spatial and textual approaches are likely to be relevant in internet research, although the balance between them will vary from project to project. The adoption of what they call ʻa hybrid model of relational ethics that incorporates text, space and bodiesʼ (Ibid, 245) enables the production of ʻresearch that examines the complex intersection of technologies, form, genre and content that the Internet supportsʼ (Ibid, 244).
Indeed, reflecting on their own research into a lesbian website, Bassett and OʼRiordan have written about their sense that they were ʻpiecing together elements from different models, none of which were entirely satisfactoryʼ (Ibid), and I also experienced such a sentiment in my own project.
Certainly I found it hard to view the content I encountered in my project only as text, without taking into consideration additional data about its creators and the broader context in which they developed, published and disseminated their content. At the same time, focusing on such practices at the expense of the texts themselves also seemed unsatisfactory given my research questions.
With such concerns in mind, then, I looked to anthropology and ethnography for approaches to texts which might help me in developing my own hybrid methodology.

The status of texts in anthropology and ethnography
Some anthropologists working on textual production in contexts unrelated to the internet have argued that texts have been neglected in their discipline, and that while they have often been used as data, they have been kept in the background and/or subordinated to other forms of data, particularly those which are oral or practice-related (Barber, 2007;Blommaert, 2008;Barton & Papen, 2010). Karin Barber (2007, 17), who works primarily on African oral poetry, compares how British social anthropology and American cultural anthropology have approached texts and finds that in the former, texts have often been treated as ʻa methodological means to an endʼ. They have been ʻsuppressedʼ (Ibid, 19) in research write-ups, and not studied in their own right. Texts have been much more prominent in the work of American anthropologists, but although their tradition can be seen as ʻan anthropology to, for, by, with and from textsʼ (Ibid), Barber does not consider it ʻan anthropology of texts, at least not in the sense of an anthropology that seeks to understand texts and textual traditions in the light of something elseʼ (Ibid, 21; emphasis in original). This ʻsomething elseʼ that Barber refers to is ʻsocial relationshipsʼ, which she places at the heart of her own anthropology of texts, asking ʻin what ways verbal textuality arises from, and in turn helps to shape, social relationshipsʼ (Ibid, 29).
Blommaert, who also works on African texts, makes similar points about how grassroots texts have been neglected and, like Barber, calls for attention to textuality itself as an important area of ethnographic analysis: Scholars, to be sure, have used such texts. But often they have insufficiently attended to the features that make such texts into what they are: products of grassroots literacy that demand close inspection of their formal features, the linguistic, stylistic and material resources that were used in them, and the various constraints that operated on this process. (Blommaert, 2008, 10;emphasis in original) Blommaert and Barber therefore unite in drawing attention to an underdeveloped area of research, or to a methodological gap or problem which needs to be addressed. Blommaert (Ibid, 12) calls for ʻan ethnography of textʼ and takes a linguistic approach, whereas Barber argues for the development of ʻa concrete, historical anthropology of textsʼ, continuing, We need an approach that can grasp together the local specificity of textual production and the larger historical forces and trends that profoundly affect without fully determining it. And to grasp the originality of the local, texts need to be seen in relation to the textual fields from which they emerge and into which they return. (Barber, 2007, 223) Like Blommaert, Barber advocates a detailed focus on texts themselves, noting that her first degree was in English, at a time when New Criticism was in vogue. As she writes, the only way to start, and the only place to end up, is with actual texts. We have to apprehend just how the words work. Too many anthropological excursions into other peopleʼs texts hover above this level of specificity -contenting themselves with summarising plots, paraphrasing prose or extracting symbols and themes from poems. (Ibid,225;emphasis in original) As these citations imply, both Barber and Blommaert focus directly on texts as the principle category of ethnographic data, and to some extent textual practices of production and circulation (or the materiality of texts, in the case of Blommaert). In my study, whilst I paid close attention to texts (or content), I also sought out and incorporated the narratives of content creators to help situate that content in a wider context, as well as observing elements of the publication and circulation of the content. In this way, I broadened my research object beyond text to also encompass practices.
In fact, practices have become an increasingly prominent focus of anthropological or ethnographic approaches to media (Bräuchler & Postill, 2010). As Nick Couldry has pointed out, a new paradigm has emerged in media research, which aims to decentre media research from the study of media texts or production structures (important though these are) and to redirect it onto the study of the open-ended range of practices focused directly or indirectly on media. This places media studies firmly within a broader sociology of action and knowledge (or if you prefer, cultural anthropology or cognitive anthropology), and sets it apart from versions of media studies formulated within the paradigm of literary criticism. (2010,(36)(37) In this way, while scholars in other areas of anthropology have been advocating greater attention to texts, the reverse approach can be discerned in certain areas of media anthropology research where there has been a move away from a textual focus, in favour of a primary focus on practices.
However, there has also recently been a direct call for greater attention to texts from within media anthropology itself. Elizabeth Bird (2010, 7) has suggested that ʻanthropologists might profitably address more attention to texts, in addition to production and receptionʼ and that by incorporating textual analysis into broader ethnographic research, they are well placed to shed light on the meaning of media texts, interpreting them in the context of their creation and reception. One recent response to this call for greater engagement with media texts in media anthropology is work by Johanna Sumiala and Minttu Tikka (2011) on school shooting videos on YouTube.

Authorship and ethics
When the media texts being studied are those authored by ʻordinary citizensʼ (Rodriguez, 2001) on the internet, particular ethical issues may arise. Authorship is an aspect of grassroots writing which is undergoing significant transformations as a result of digital technologies. As Barber (2007, 222) notes, ʻopportunities for individuals to project their personal writings into public space have multiplied, with desktop publishing and the rise of the blog. Authorship has proliferated while the aura of author as cultural originator has dispersedʼ. Authorship is a key concern of literary and cultural approaches, but must also be addressed in approaches which include the use of research methods from the social sciences. Whilst research ethics are not the core focus of this article, it is important to point out that a key issue faced by researchers working with digital texts produced by research participants is the relatively easy ʻtraceabilityʼ (Beaulieu & Estalella, 2012, 10) of such texts if verbatim quotes are provided in research write-ups. A decision is therefore also usually necessary about whether research participants are to be considered authors, human subjects, oras often happens -, a hybrid of the two (Bassett & OʼRiordan, 2002;Bruckman, 2002). Although it has been noted that current approaches to anonymisation in internet research are often too thin, iv some scholars have argued that not citing from internet content on principle would cause significant challenges, and even make unviable, research into discourse and rhetoric on the internet and its subsequent publication (Hine, 2000;Banks & Eble, 2007). A further consideration is whether the content has been published to counter some kind of marginalisation or media invisibility, and how its inclusion, or omission, in research would affect this (Bassett & OʼRiordan, 2002;Banks & Eble, 2007). Overall, researchers need to take decisions on such issues in the context of their own research and ideally find solutions in consultation with research participants (Hine, 2000;Franklin, 2004;McKee & Porter, 2009).
Beyond the ethical considerations, the far-reaching methodological (and indeed theoretical) questions raised by such media texts are highlighted by Eichhorn, based on her work with ʼzines: Can ʻtextsʼ that promote the sort of interactivity and immersion experienced with ʼzines, as well as some forms of hypertext, continue to be understood as texts? At what point do these ʻtextsʼ become something entirely different? If so, what do they become, and what might the phenomena in question mean to the ethnographer, the literary theorist, and their respective disciplinary traditions? (Eichhorn, 2001, 576) With such (inter)disciplinary dilemmas in mind, Bassett and OʼRiordan (2002) provide a brief but useful overview of existing scholarly approaches to the relationship between (mainly print) text and author (text as a direct reflection of the authorʼs values; text as object to be analysed independently of its author, readers or the conditions of its production; and the text viewed principally through the lens of the readerʼs response and the meaning this creates). They find more fertile ground for an ethical approach to personal texts, one which takes into consideration the role of the author, in the field of Life Writing and particularly the work of Katherine Borland. Although Borland (1991, 73 cited in Bassett & OʼRiordan, 2002, 240) advocates a sensitive approach, given the link between such texts and real people, she does not conflate authors with their textual narratives; her approach involves an exchange between the researcher and the author of the text, but also engagement with the text itself by the researcher, and I attempted something similar in my own research, combining my own textual and visual analysis of internet content with data collected through interaction and interviews, which sometimes involved direct discussion of specific instances of content with its creators.
As this review of approaches to texts has shown, areas of ambiguity persist in relation to the methodological and ethical status of texts in anthropology, ethnography and internet studies, as well as in relation to issues associated specifically with digital texts. However, the authors cited above offer concrete ideas regarding how texts can be tackled within these fields, and combined with other types of data. The final section of this article discusses how the methodological concept of the ʻcontent eventʼ, inspired by new literacy studies, was employed in my project as a further way of linking texts and practices.

Bringing together texts and practices: Borrowing from new literacy studies
New literacy studies is a field of research which examines the role of reading and writing in society from an often multidisciplinary, and primarily ethnographic, perspective (Barton & Papen, 2010, 11). It focuses on largely everyday, vernacular texts, traditionally unpublished or unbroadcast in the conventional sense, although this characteristic may be changing with the internet. However, despite the interest in texts, practices have also tended to dominate the analysis here to a certain extent, with the focus being ʻhow written texts fit into the practices of peopleʼs lives, rather than the other way aroundʼ (Barton & Hamilton, 1998, 257). Nonetheless, based on their study of literacy practices in Lancaster, northern England, David Barton and Mary Hamilton (1998) concluded that many of the texts encountered in the course of the research were interesting in their own right, and deserving of closer attention. As they argued, there is potentially rich data to be ʻgenerated by a strategy of identifying significant texts from a study of practices and moving between the analysis of texts and practices in a cyclical way to develop an understanding of contemporary literaciesʼ (Ibid,258). Similarly, the editors of a recent collection on the anthropology of writing, which includes work in new literacy studies, argue that it is often possible to connect ʻordinaryʼ texts, part of ordinary life, to ʻbroad, complex and at times extraordinary social eventsʼ, and that this field of research therefore touches on ʻissues that are at the heart of contemporary anthropology: knowledge and power, identity, social change and the interface between local and global spacesʼ (Barton & Papen, 2010, 10). Such issues were present in my own study as I explored how favela bloggers were using the internet to affirm and reposition their neighbourhoods as an integral part of the city in contemporary Brazil, and negotiating complex questions associated with the locality and visibility of content published on the internet.
Reflecting the resonance of core concerns from new literacy studies in my own research project, then, it was this field that provided me with a specific methodological concept, the ʻliteracy eventʼ, which became the ʻcontent eventʼ as I adapted it to the context of my research. The term ʻliteracy eventʼ, which originates in the work of American linguistic anthropologist Shirley Brice Heath, is employed in literacy studies to refer to ʻobservable episodes which arise from practices and are shaped by themʼ and often have written texts at their heart (Barton & Hamilton, 1998, 7). In their influential interpretation of the term, Barton  can also be more exceptional and dramatic. Inspired by this concept, then, I developed the concept of the ʻcontent eventʼ as a way to connect local content texts and the practices involved in their production, publication and dissemination across different platforms. In my study, I considered three empirical cases, each of which incorporated a content event. The first related to a single text written in response to a local conflict, which was originally published on a group blog and then reposted on other blogs and websites by its author and others, as well as being disseminated on Twitter, social network sites, email and in print, undergoing some changes in the process. The second content event encompassed the development and promotion, through the internet and other channels, of a writing competition for residents of the area, which evolved into a broader competition open to writers from suburban and periphery neighbourhoods more broadly speaking.
The third content event incorporated thematic content clusters or ʻseriesʼ, including texts, photographs and videos, published over a period of time by an individual blogger. In my analysis of these content events, I incorporated my own observations, close textual and visual analysis of content, and the narratives of the content creators in question, ʻin order to illuminate relations between digital texts and their production and reception practicesʼ (Androutsopoulos, 2008, n.p.).
From early in the research process, then, I established my intention to be free to follow links and connections, to accompany local content as a text and as a practice as it circulated, not necessarily in its original or published form, between users, between platforms and between locations, as well as the flows of new content and changes to ʻframing contentʼ. I paid increasing attention to how particular stories, messages or announcements were published, disseminated on different platforms, and sometimes reposted. Wherever possible, I decided to follow such links and connections, exploring the way that content travelled with or without the direct participation of its original creator. I therefore did not have to go (only) to a particular favela to find my field. I found it (principally) on the internet and in local content as published and disseminated across different websites and platforms, content that then took me closer to areas of the favela and aspects of favela life that I did not experience for myself. Interviews with content creators also fulfilled the same purpose. But I also found my field (or followed the contours and extensions of my field) in other areas of the city of Rio, for example when I attended a book launch in a suburb of northern Rio, organised by a blogger with whom I had corresponded by email before meeting her in person on a previous occasion, or when I attended a central Rio screening of a film, set in the favela, which was followed by a discussion panel involving local residents. I also encountered my field in local newspapers and on the websites of mainstream media organisations, sometimes seeing the same stories covered in content published by residents, or hearing them discussed in interviews.
However, there were often lulls during fieldwork, leading to a sense of ʻwaiting for something to happenʼ (or be published), as I checked feeds and sites at my computer, looking for relevant content. Given that the flows of content from particular sites were often not substantial, intense or predictable, it proved productive to reorient my efforts to focus on the evolution of content sites, the connections between content items and sites, and to take an in-depth look at the content itself. In this way, my dual strategy proved fruitful: by following the content produced by specific local content creators, I was also following their practices. Establishing mobility as a key principle of the fieldwork from the outset, despite my focus on place, was a response to the connective nature of the internet (and of place itself), and enabled me to be attentive to the way local content, and the practices of local content creators, traversed and occupied different internet platforms and communication channels, as well as different places within the city and beyond it (which were also represented in the content). This provided me with an awareness of the fluidity and multiplicity of local content, and locality, which might have escaped me had I chosen to site my fieldwork more exclusively in a particular physical location, or not sought to connect local content with the practices involved in its production, publication and dissemination.

Conclusion
This is not intended to be an exhaustive overview of existing approaches which link texts and practices, but it presents reflections and solutions inspired by the methodological questions and issues which arose in my interdisciplinary study. Drawing on the diverse readings I engaged with in the process, from anthropology, internet ethnography, new literacy studies and internet studies, I have discussed how I adapted and built on existing approaches in the context of my research.
Indeed, although my own ʻpiecing together [of] elements from different modelsʼ (Bassett & OʼRiordan, 2002, 244) was sometimes challenging as I sought to bring together diverse references, influences and approaches in the theoretical and methodological shaping of my interdisciplinary project, it was also productive and creative.
As I have shown in this article, the ʻfollowing the contentʼ approach I employed was an enactment of multi-sited ethnography which responded to mutable and mobile sites and flows of local content.
This mode of research engagement with user-generated content also acknowledged the necessarily ʻpartialʼ (Hine, 2000, 10) nature of my ethnographically inspired study, given the primary focus on the local content outputs of specific individuals from one favela in Rio de Janeiro, and the way I constructed my field through engagement with that content and the links and connections it offered me. Although this framing of the methodological practice emphasises the focus on content, or text, I have discussed how the alternative of ʻfollowing the peopleʼ persisted implicitly within the research, given the parallel interest in the practices of content creators and the attempt to bring together these two types of data. Ultimately, ʻfollowing the contentʼ was both a conceptual device for thinking about the fieldwork process, and a description of the practicalities and logistics of the research process. Similarly, ʻcontent eventsʼ, borrowed and adapted from new literacy studies, offered both a conceptual and a concrete way of linking content to the practices involved in its production, dissemination and sometimes reception, and organising the data into manageable units for analysis.
When it came to analysing the research data, I was able to cross-reference material collected through interviews and observations with textual and visual analysis of content, and again this was crucial in developing an in-depth contextual understanding of local content in a way that brought together texts and practices. In this way, the interdisciplinary approach used drew on calls for greater attention to texts in anthropology and ethnography, alongside the increased prominence of practices in these fields, as well as reflections by other internet researchers who have attempted, like me, to develop a hybrid approach linking texts and practices, or content and people.  (Sorj & Guedes, 2005, 4-5, 9). The national figure for 2010 was 41% (CETIC.br, 2011b). If the same correlation has been maintained, therefore, the current level of internet access in Rio favelas is also likely to have risen significantly since 2003. Indeed, in 2012, the Fundação Getúlio Vargas, a Brazilian research institution, released data showing that whilst large favelas in Rio had the lowest levels of digital inclusion (understood in the study as access to mobile and fixed telephony, and a computer with internet at home) in the municipality, levels of access in those areas were still higher than the national and global averages (Quaino, 2012).