Ethics of Internet Research: Contesting the Human Subjects Research Model

E. H. Bassett
School of Information Management
Watts Building
University of Brighton
Lewes Road
Moulsecoomb
Brighton
BN2 4GJ UK

Kathleen O'Riordan
Lecturer in Continuing Education (Media Studies)
Arts C304
CCE
University of Sussex
Falmer
Brighton
BN1 9RG UK


he human subjects research model is increasingly invoked in discussions of ethics for Internet research. Here we seek to question the widespread application of this model, critiquing it through the two themes of space and textual form. Drawing on our experience of a previous piece of research (R. Munt, Basset & O'Riordan, 2002), we highlight the implications of re-considering the textuality of the Internet in addition to the spatial metaphors that are more commonly deployed to describe Internet activity. We argue that the use of spatial metaphors in descriptions of the Internet has shaped the adoption of the human subjects research model. Whilst this model is appropriate in some areas of Internet research such as email communication, we feel that researchers, when navigating the complex terrain of Internet research ethics, need also to consider the Internet as cultural production of texts. 

Introduction

Internet research is carried out across a broad range of academic disciplines, including media studies, cultural studies, literature, studies of the arts and sociology. Most of these examples share a concern with the sociocultural and examine the behaviours and expressions of human actors in a social context. However, as Herring asserts (Herring, 1996), the complexity of Internet research cannot be anticipated and should not be prescribed. Some research environments are experimental, clinical or controlled (Bruckman, 1998); other environments support participant observation, or textual analyses. Some anthropological and sociological studies combine research methods from a range of different fields. (Kendall, 1996; Danet, 2001, & Senft, 2002)

The ethics of Internet research have become a significant area of concern as use of the Internet, as research object and medium, has increased. Examples of the variety of Internet research include 'virtual ethnography (Markham, 2000), linguistic and discourse based analyses of computer-mediated communication (Herring, 1996; Sharf, 1999; Baym, 2000; & Danet, 2001) and research into the Internet as culture. (Wakeford, 1996; Clerc, 1996; Hills, 2001; Miller and Arnold, 2001; Sveningsson, 2001; & Senft, 2002). A wide range of research, theoretical, empirical, directed, grounded, qualitative and quantitative, has been pursued in these contexts, with Internet research to date being characterised by a diversity of methods, objects and perspectives (Bell & Kennedy, 1999). The extensive discussions of research ethics have provoked the development of very different guidelines from a number of sources, and we largely concur here with Herring's 'social realism' (Herring, 1996a) whilst discussing a different kind of analysis and emphasising the interplay between space and text. 

The perception that the Internet is a social domain has led to the widespread application of the human subjects research model in Internet research. Human subjects research ethics, which emerged from the post-war view of research, regard the rights of the human subject as primary and the aims of the researcher as secondary. There is a strong argument for the use of this model when research is conducted into activity such as email between individuals and groups, for example Sharf (Sharf, 1999). Case studies in the literature that evidence harm to the researched, referred to by King (1996) and Turkle (1996), have fuelled an imperative to apply restrictions drawn from the human subjects model. It is particularly applicable when groups and individuals use computer-mediated communication to convey 'private' material through channels that have some mechanisms to limit access. However, when applied indiscriminately, this model risks subsuming the human subject into the technology and eliding the textuality of the Internet. In a climate of fast developing technologies, it is often difficult to define the actual object of Internet research precisely. The human subjects model is more often than not considered to be the 'safest' model to apply. We argue here however, that the cost of this 'blanket - approach' may be a resulting failure to analyse the significant textuality of Internet media.  

Internet 'Space'

Two influential lines of thought have led to the application of the human subjects research model to Internet Research. The first is the dominant metaphor of spatiality that is applied to the Internet. The use of space as a way of understanding the Internet is widespread, and perhaps inseparable from fundamental aspects of human cognition:

Human knowledge passes through two forms of cognition before it can be conceived: space and time …in other words we know that, before we experience things, we will perceive them as phenomena in space and time. (Munt, 2001, p.2.)

The understanding of the Internet-as-a-space supports a conflation between activity carried out through this medium and the action of human actors in social space. Further, it leads to the argument that any manifestation of Internet activity should be regarded as a virtual person. This conflation of the spatial metaphor with the technology is helpful up to a point but is ultimately limited; indeed, adopting the spatial metaphor exclusively, we will argue, can sometimes lead to unethical results. The Internet is not simply a virtual space in which human actors can be observed: it is a medium through which a wide variety of statements are produced. Thorough research requires that the Internet technologies as media be differentiated according to a variety of factors such as content, genre and form. Subsuming the heterogeneity of the Internet to a homogenous whole is a reductive move. Furthermore, it risks making the unsupportable conflation of the Internet user with their textual output.  

The second metaphor that underpins the construction of Internet technologies as social space is virtuality (Michael Benedict, 1995). The slippage between real and virtual has been invoked in arguments that expressions of the self and simulations of persons made through Internet and computing technologies, can be translated as the self or person who authored them. This conflation of person and text is also influenced by the model of the body as data, which emerged in the 1960s (Hayles, 1999), and is sustained through cybernetic, cyberpunk, extropian and other technologised subcultures (Terranova, 2000). The transhumanist Pepperell (1995), the roboticist Moravec (1999) and the cyberneticist Warwick (1999) have all argued that the emotion, feelings and sentience of persons can be replicated and translated through computing technologies:

...computers will eventually be capable of the same kind of perception, cognition and thought as humans. (Moravec, 1999)

There is a paradigmatic link between these arguments, drawn from the biomedical and behavioural sciences, and the human subjects research model; that of the conflation of the metaphor for the object it describes. This equivalence or synechdocal logic allows a slippage between the human nervous system and technology, between person and their textual manifestation. This is more familiar as the theological paradigm of the translation of forms (re-incarnation, ascension). The same paradigm that allows a vocabulary of transubstantiation to emerge is renewed in the context of the cultural construction of the Internet as a 'new' space. So Wertheim observes that

In our time of social and environmental disintegration …today's proselytizers of cyberspace proffer their domain as an idealized realm 'above' and 'beyond' the problems of the troubled material world. (Margaret Wertheim, 1999, p. 16)

Wertheim's discussion of the spatiality of cyberspace emphasizes how this 'synecdochical logic' of the spatial metaphor falls back on specifically theological assumptions. This framework of a dualistic reality, which many users of the same metaphor would reject, underpins this 'space'. 

Many Internet content providers develop the use of spatial metaphors in the branding and packaging of their products. The basic, although metaphorical, unit of Internet 'space' is a case in point. Regardless of brand name, a generic 'web space' is used by Internet Service Providers (ISPs) as the product that the Internet user purchases along with email and Internet access. The denotation of web pages as 'Homesteads' is another explicit marketing strategy that is dependent on the spatial metaphor to conflate web page hosting with land and property. The homestead metaphor is drawn from Rheingold (1993), and is an example of the link between theoretical critiques of the Internet and the commercial market. The language used in these two spheres shares a vocabulary of the spatiality, speed and newness of Internet technologies.  

The Human Subjects Research Model

The American Association of the Advancement of Sciences (AAAS) report (Frankel & Siang, 1999), written to inform Internet research, provides a valuable mapping of some of the key issues and points to how the human subjects research model can be adapted. Yet the report appears to conflate human subjects and the Internet technologies that they use, with the consequent elision of Internet textuality. The report refers to the Internet exclusively as a 'site' of community and interpersonal interaction without greater elaboration of what is meant by 'Internet'. This reference to all aspects of the Internet as a 'site', invoking space as the primary paradigm, implies that all Internet research is human subjects research. This is a severely reductive view of the Internet. The Internet consists of a heterogeneous array of different media; the diversity of content that this media supports accrues to far more than interpersonal communication in spatial realms. The Internet-as- (community) -space model is perhaps a historical production of the pre-web Internet, yet it must be acknowledged that even the pre-web Internet had its cultural artifacts and public debate as well as inter-'personal' communication. There is a need for an acknowledgement of the textuality of the Internet, with Internet technologies now being used to publish texts produced in a complex convergence of written text and audio/visual media. 

Internet Textuality

The Internet is not only a text-based medium made up of communities, newsgroups and email lists. It is also a medium of publication, and significantly one where users can take control of the means of production, create their own cultural artifacts and intervene in the production of existing ones. The Internet can thus be perceived as a form of cultural production, in a similar framework to that of the print media, broadcast television and radio. The Independent Media Centre (IndyMedia.org) web sites are a case in point. These web sites publish 'news' items written by activists from all over the world. The sites were established with the aim to contest mainstream news production, to write alternative histories and to validate and draw support for alternative politics. A decision that this centre could not be discussed within the academic sphere without recourse to the human subjects model would undermine these aims.  

The Independent Media Sites are made up of postings in CMC formats including traditional text based email, bulletin board and chat technologies. Some 'regulars' create posts but many people only place messages as and when they are involved in a political issue. Many of those using the site are at risk from people around them and 'harm' in this context could mean fatality. However, this is a political forum in which activists choose to make visible the political events in which they feel involved or which they consider important. They also make a decision about the level of their visibility by controlling the degree of their disclosure of identification. There is no sense in which the Independent Media Sites are private although they are highly 'sensitive'. They constitute a deliberate attempt to create a global public sphere through Internet use. Most of the work is not published in a traditional sense, nor is it copyrighted. Many sites are largely text based although some also include visual media. However, the application of the human subjects model to the site could be enforced as a condition of research under the USA system of Institutional Review Boards supported by the Frankel and Siang report (Frankel & Siang, 1999). As we have just attempted to argue, there are a number of reasons for not treating this work as private and thus requiring deontological human subjects protections such as anonymity, seeking informed consent, etc. In addition, we would note an important consequence of applying the human subjects model: Gaining consent from all participants and evaluating risk and benefit, in the global and mass context of Independent Media would in our view be counter productive and impossible, except under the most well funded conditions.

In addition, an adoption of the human subjects model risks academic censorship of the news media produced on the IndyMedia web site. Nobody would suggest that the text of a news item published in a newspaper be conflated with its author and considered as a human subject, and yet this is exactly what is happening regarding Internet texts. This different valencing of online and offline texts could result in academic research contributing to an uneven power relation, which validates print newspapers produced by major international companies as legitimate objects of study. This would obstruct and ultimately prevent the research of texts published on the Internet, many of which, like those on the IndyMedia web site, are published by organisations lacking the resources to compete with the print and broadcast media conglomerates. We argue in the discussion of our case study that, just as applying the human subjects model to the IndyMedia.org site would end up undermining its goals, so in other more 'traditional' arenas of CMC, applying conditions of the human subjects model, can likewise be counter productive. 

Textuality

Textual metaphors, although not as widely applied as spatiality at present, have also been used to describe networked communication, with the hypertext metaphor persisting from the writing of Vannavar Bush (Bush, 1945, pp. 47-61) through that of George Landow (1997) to the present. Academics such as Hayles (1999) and Bassett (2001) contest the dominance of a spatial paradigm by emphasising narrative as a way of understanding the Internet. This emphasis on textuality can also be read in the concerns over Internet copyright. Viewing Internet content as a text, as a culturally produced object, invokes the concepts of ownership and an economic index (Cavazos, 1994). The complexity of this issue affects content producers and critics alike, and is evident in the focus placed on MP3 consumption by the Recording Industry Association of America. Yet, despite its complexity, this understanding of the Internet as textual must be given greater consideration. The increasing trend towards conflation of Internet media and space, such as the denotation of the Internet as a 'site' in the Frankel and Siang report (Frankel and Siang, 1999), make a reconsideration of the textual nature of the Internet ever more pertinent. Issues of privacy, confidentiality and informed consent remain central to research considerations. Yet we must also address the textual status of the Internet, and consider issues such as appropriation, reproduction and removal of Internet texts from their original context.1

There are many different ways in which texts have been conceptualised, and key to the consideration of texts of any kind is the relationship that they have with the person or persons who authored them. The question of what a text reveals about the writer has resonated throughout the history of the discipline of Literary Criticism. There are three main models that are briefly described here: text as reflection of the author; text as object; and text as reader-response. 

One model is literature as a textual opening into the mind of the revered author (text as reflection). As literary critic and theorist Terry Eagleton put it, proponents of this view such as I. A. Richards felt that

Great literature is the product of Great Men, and its value lies chiefly in allowing us intimate access to their souls… (Eagleton, 1983, p. 41)

The text was considered to be a faithful reproduction of the values the author held in (his) soul or mind. (This perspective included an assumption that these were minds contained in male bodies). The relevance of this viewpoint is that the text is all but conflated with the mind or indeed the soul of the author. This is a 'mind' made word, or text as a precursor to the virtual self. The logical extension of this conflation of text and mind is that whatever respect is accorded to the minds of these 'Great Men' should be accorded to their textual output. The reverence bestowed by academia on the literary canon in the early twentieth century is an indication of the widespread influence of this viewpoint.  

This conflation of text and author was challenged throughout the twentieth century, although the model is still used. The school of New Criticism, for example, was highly influential in Northern American and, to a lesser extent, in British Literary Criticism in the post-war period. Proponents of New Criticism sought to sever the text from both the author and the reader (Eagleton, 1983, p. 41). New Criticism was concerned with the interpretation of texts. Like semiotics, it held that whatever the intentions of the author had been at the point of writing, the meaning of the text lay solely in the language used. The text, (and New Criticism was concerned primarily with poetry), was considered to be an organic whole, and no more a window into the soul of the author than into the soul of whoever was reading it. The notion of attempting to find the author in the work, or vice versa, was labeled the 'Intentional Fallacy' (Wimsatt & Beardsley, 1954). The text, now rendered a 'self-sufficient object', was also liberated from both social and historical context of production (Eagleton, 1983, p. 42). With New Criticism the text-as-virtual-writer is replaced by the text-as-object, severed from context, authorial intent, and the response of the reader.  

The assumption that a piece of text is an object underlies the logic of copyright law. The complexity of this issue and the developments in copyright in relation to written texts was highlighted at the end of the 19th century by debates over the works of the novelist Charles Dickens, for example.2 Thus, issues of ownership, space, dissemination and meaning have been debated in relation to literary texts over the last two centuries. Subsequent copyright law is based upon property law and works on the basis that a text is 'intellectual property', belonging to an author in the same way as any other object s/he produced. This model of ownership is related to a shift in the political economy of production, which changed from the patronage system (royal or noble sponsorship), to the recognition of the individual as cultural producer. In copyright law the text is not a window onto the author's soul but more akin to a box. The contents belong to the author and s/he is free to distribute them and adapt them as s/he wishes. The author sells the distribution of the text, either in units, receiving payment each time the text is reproduced, or outright, by signing the authorial rights over to a publisher for a certain length of time. The reader, by buying a copy of the book, purchases the right to read the work but does not acquire the right to redistribute it. The author has a right to be recognised as the author of the text, but this means recognition of the fact that s/he owns the text and produced it, not that the text reflects her point of view. 

Copyright extends to written and recorded words, musical composition, sound and film recordings amongst others. It has more recently been applied to items on the Internet such as news stories, software, novels, screenplays, graphics, and email. Many of these are covered under existing copyright law, which is as applicable to the Internet as it is to any other form. Texts that are not covered by copyright include those that did not contain a valid copyright notice, and those for which the copyright period has expired. These texts are considered to be in the public domain, alongside any texts that the author has granted to the public domain. The registration of copyright of texts distributed via the Internet counters the notion that the Internet is a purely spatial realm. The dissemination of text through Internet technologies has also strengthened the 'copyleft' movement, which seeks to enable 'free' dissemination. Although posing a challenge to copyright, 'copyleft' is simply the inverse of the same paradigm of ownership. Stallman (1999), details this in 'The GNU Operating System and the Free Software Movement', but as well as relating to software and 'warez' (Tetzlaf, 2000) copyleft has been taken up by activists disseminating text (e.g. IndyMedia and Schnews (UK)).  

When it comes to the practice of registering domain names textual and spatial characterisations of the Internet are invoked. A domain name can be an equivalent to a trademark, but it is also a unique alias for an IP address, which relates to an actual physical point. A domain name is therefore a piece of text that is both a piece of property covered by copyright law and a signifier of a location. This double identity is evident in domain name disputes where entrepreneurs register the names of famous people or companies in the hope of making a profit by later selling the registered domain name to the relevant party. Although this practice is clearly a sidestepping of copyright law, and therefore involves the appropriation of property, the metaphorical spatiality of the Internet is reinforced through the informal term for this practice: 'cyber-squatting'. 

If copyright treats a text as a piece of property belonging to, yet distinct from the author, contemporary schools of literary criticism agree that the relationship between the meaning of a text and its reader is less clear. Reader -response theories, and phenomenological hermeneutics both hold that the text separates the author from the reader (Lye, 1996). As philosopher Paul Ricoeur states:

…the book divides the act of writing and the act of reading into two sides, between which there is no communication. The reader is absent from the act of writing; the writer is absent from the act of reading. The text thus produces a double - eclipse of the reader and the writer. (Ricoeur, 1982)

Riceour's hermeneutic theory can be compared to the deconstructionist 'Death of the Author' model, which deconstructs the notion of a text as the product of an intentional author in possession of the meaning (Barthes, 1977). Foucault's (1972) work locates issues of meaning and agency in relation to language as part of a wider discursive formation, adding further complexity to the writer-text-reader relationship, and destabilising further models of authorial or communicative 'intention'. The resulting separation between author, text and reader prompts the question of to what degree a text can be said to have any intentional or reflective meaning. Critics of many theoretical persuasions, particularly since the deconstructionist movement of the 1960s, agree with the tenet of reader-response criticism that:

…[as] the meanings of a text are the 'production' or 'creation', of the individual reader, …there is no one correct meaning, either of the linguistic parts or of the artistic whole of the text. (Abrams, 1941, p. 150)

The notion that the meaning of a text is brought about in the process of reading adds different levels of concern for those involved in researching texts. The contemporary text now appears to be both permeable and malleable. The notion that the researcher constitutes the text contests the principles of objective and positivist research. In these paradigms the role of the researcher is to uncover the 'truth' of the object of study. This principle does not only apply to the study of inanimate objects: it is a traditional tenet of anthropological research that the researcher attempt to leave the object of enquiry unchanged. 

Contemporary research perspectives, along with literary theory, acknowledge the impossibility of conducting research without altering the meaning of what is observed. Feminist research in the 1990's in particular promoted the need for the researcher to account for their influence on their object of study (Stanley, 1995). This view holds that the meaning of any situation or text does not exist prior to observation, but is inscribed in the process of research. This understanding has lead to the use of reflective and 'participant-observation' methodologies (Mayer, 2001) that clearly convey the participatory role of the researcher. 'Post-paradigm' anthropology (Marcus & Fischer, 1986) and feminist sociological research both require the researcher to account for the dynamics of the relationships between researched, researcher and results (Stanley, 1990; Skeggs, 1995; and Mayer, 2001). Feminist documentary and film making has also stressed the relationship between the observed or filmed and the producer of the text (Minh-Ha Trinh, 1991). In these paradigms the text is a contested and multiple output with a multiplicity of stakeholders. To make this point another way, the implication for the social construction of knowledge through research is that it is a contested and relational construct (Harding, 1992).

When it comes to researching written texts, ethical guidelines are not easy to locate, except within linguistics. Whereas a degree in Sociology will routinely require the student to engage with the debates in ethical practice, colleagues enrolled on a Literature course will not necessarily be asked to consider their relationship to texts from an ethical standpoint. A sharp division is drawn between ethical considerations required when researching people, and when researching texts. This contrast is striking in responses to questions around privacy and the Internet. Debate is polarised - for example - ask whether it's ok to quote a dialogue you copied from a chat room and the response you'll get will follow one of two possible routes. One view is that a chat room is a space and the dialogue is between people in it, all of whom are due the protection accorded them under the human subjects research model. There are variations on this that consider what kind of space it is, how private it is, and how much protection the virtual subjects should be accorded, but the dynamic is the same. The second view often takes a more direct route; the dialogue you have is a text, it's in the public domain, and therefore, aside from considerations of copyright, is available for reproduction. This attitude is reinforced by journalistic methods of representing the Internet, which sometimes reprint or create 'spoof' chat rooms (Dowling, 2002).

It is not necessary that the ethical debate around Internet research revolve around a polarity between space and text. If we want to consider a more inclusive approach then there are leads to be found in academic disciplines that engage with texts offline and the people who produce them. The study of Life Writing spans many disciplines, and represents 'a fast changing terrain' which is concerned with the study of the writing of lives, encompassing auto/biography as well as aspects of life story that originate outside the written form, such as oral history, testimony, and artifacts such as photography and the visual arts (Jolly, 2001, p. ix). The written text is viewed as a vehicle for an individual's construction of identity, the personal narrative that it records being

…a fundamental means by which people comprehend their own lives and present a 'self' to their audience. (Borland, 1991, p. 71)

Despite the symbiotic relationship between text and personal identity, Life Writing practitioner Katherine Borland does not conflate the text with the person who authored it. Rather than claim that the written narrative is a virtual self, and therefore due equivalent ethical consideration as the real life author, Borland instead seeks to explore ways of dealing with texts responsibly (Borland, 1991, p. 73). The approach she advocates acknowledges the direct relationship between these texts, and research of them, with their author. Borland states that:

…scholarly representations of those performances, if not sensitively presented, may constitute an attack on our collaborator's carefully constructed sense of self. (Borland, 1991, p. 71)

Yet, Borland does not claim that this potential danger necessitates a shift into a human subjects model. The textual narrative, whilst linked to the author, is not conflated with them. Borland calls for a more open exchange of ideas between the Life Writing researcher and the author of the text, person to person, but she also reinforces the value of the researcher's engagement with the text itself (Borland, 1991, p. 73).  

The multiple approaches taken in Life Writing research can arguably be attributed to the multidisciplinary background of its practitioners. Like Internet researchers, those researching Life Writing have been trained in a variety of disciplines including anthropology, sociology, history, linguistics and literary theory. These various disciplines have contributed a range of insights into the collection of oral history and analysis of the resultant texts (Gluck and Patai, 1991, p. 3). Ethical guidelines for the conduct of Life Writing researchers, such as those published by the Oral History Society in the UK, appear to follow a human subjects model, focussing on the undertaking of the interview, and restricting discussion of the resultant text to issues of copyright. Yet practitioners emphasise the need to remember that as:

…the typical product of an interview is a text, not a reproduction of reality …models of textuality [are] therefore needed. (Gluck and Patai, 1991, p. 3)

Case Study

We review here a piece of research that we carried out in 2000 that considered both spatial and textual models of the Internet. Describing the considerations we made at the time of writing up we discuss the changes that we would make in future work. Although we believe that we carried out a responsible piece of research and that we made appropriate decisions, we also pursued the 'safest' option and we were constrained by the conditions of publication - namely, peer review and editorial advice. Given the opportunity to reflect on our work we would now make a much stronger claim for disclosure of the web site analysed. We do not claim that we should retrospectively change the methods of writing up. However, in future work we would not automatically follow this safest option. We further argue that the spatial framework that supports the human subjects model does not apply to cultural production of this kind. 

The web site analysed was a self-defined community/magazine that was rhetorically constructed through spatial metaphors and lesbian signifiers (Munt, Bassett, and O'Riordan, 2002). With the exception of a few web sites, for example those which are described as 'graffiti walls', there is a lack of familiar vocabulary to describe web sites in non-spatial ways. We referred to the web site through the pseudonym Gaygirls.com. We decided not to disclose the actual name of the web site primarily because of the way participants used the particular section that we analysed. Although clearly in the public domain, the participants' use of this section indicated that they perceived it as a semi-private space. They used confessional postings and stratified their audiences by discussing 'other' audiences to whom they would not communicate this information, thus implying that they had specific assumptions about who would use the web site. We also used pseudonyms to refer to individual participants when discussing their use of language. However, as well as taking into consideration how the participants used Gaygirls.com we also made this decision at the time because it was the 'safest option'.  

About The Forum and The Research

Gaygirls.com had a strong textual character, following a widely used magazine format with articles, news, images, gossip, star signs, book, film and music reviews. Like many commercial 'community' web sites there was a recognisable form: a front 'page' with internal and external links. The internal pages contained popular cultural phenomena related to the concept of the lesbian community; these included coverage of lesbian and gay film festivals, Pride events, and lesbian associated celebrities. Relationship and dating advice were all predicated on same sex relationships between women. Identity signifiers such as lesbian, dyke, amazon, LGBT (lesbian, gay, bisexual and transgender) were all used on the front page and throughout the web site. The tone was international with a strong bias towards the use of North American cultural signifiers.

The original analysis focused specifically on the use of a bulletin board, which the participants used as an online discussion forum. We use the term 'forum' to refer to this specific area of analysis and 'web site' to refer to Gaygirls.com as a whole. The web site itself was well researched and specifically marketed towards a lesbian subculture. It was also linked to networks and web rings with an established reputation for having quality content and a 'genuine' interest in their targeted demographic. However, more significantly, the web site clearly facilitated participants who engaged supportively, knowledgeably and remained focused and on topic. The threads had longitudinal existence, carrying on over a period of months, and participants were using the web site repeatedly. An active community forum was engendered through a variety of techniques. 

It is important to note the role of narrative and textuality on this site. The deployment of a mythical 'community status' on the web site as a means of eliciting participation from people who identify as LGBT is meaningful as it resonates with a central narrative form in the construction of lesbian identity, the coming out story. This narrative form describes an individual's journeying towards an imagined community, and is part factual and part metaphorical. As Sally R. Munt describes it:

The classic lesbian journey, in the coming out story, is from isolation to inclusion, an idealized trajectory of self-realization. It follows the footpath of all those hundreds of thousands of Westerners since World War I who made the geographical shift - the heroic journey - to the gay and lesbian homeland, the city. The imagination and experience dialectically constitute the lesbian subject. (Munt, 1998, p. 174)

The Textual Spatiality of the Forum

Computer Mediated Communication (CMC) is often conceptualised in terms of spatial metaphors such as those of going to, visiting, and being in or out of a forum. Web based forums evoke the notion of a particular location through the metaphor of the web 'site'. Gaygirls.com reinforced this through the repeated use of 'place' and other spatial descriptors in the rhetoric. Commercial but open access domains such as these require consideration because although they are a 'public' space in both the Habermasian senses of the literary and the political (Seidman, 1989), they also engender intimate forms of communication. The language used was personal, informal and intimate in content. There was an illusory sense of partial privacy because the participants constructed utterances that they stated they would not convey to certain audiences such as their family. This facilitated the participant's illusion that Gaygirls.com was a space over which they exercised some control, and in which they could expect quite high levels of confidentiality, safety and freedom. In contrast to the participant's expectations, there was, in actual fact no control over who viewed the material on the web site: forums on Gaygirls.com are in no way 'private'. The language used by the producers of Gaygirls.com indicated that the forum was open and public and they likened it, through simile, to a way of amplifying voices in public debate.  

When we were writing up the research we borrowed from Stine Gotved's (2000) discussion of 'parochial space' to reflect the way that participants conceptualised Gaygirls.com. Likened to a neighbourhood or residency association controlled areas, parochial space is familiar and certain levels of privacy and safety are engendered. However, it is still open to social forces beyond the direct control of the members. This parochiality is also a useful concept to describe the 'space' of the lesbian community, which is also presumed to be known and self-regulating but is also open to intervention by social forces beyond the control of its members. The illusion of control allows participants greater freedoms, acting like the literary and dramatic strategy of 'the suspension of disbelief'. In this paradigm, decisions as to how the self defines reality at a given time construct the subject at that moment: 'The imagination and experience dialectically constitute the lesbian subject' (Munt, 1998, p. 174).

There was considerable variation in the ways that participants, owners, and we as researchers of Gaygirls.com perceived it. We viewed it as a form of textual production, whereas the web site organisers presented it as textual, but also as spatial and public. The participants' use of the web site suggested that they viewed it as spatial and 'parochial'. Gaygirls.com was represented as a 'safe' environment through explicit textual statements by the participants, but not by the producers. Ten percent of the contributors directly described either the Internet in general or the Gaygirls.com in particular, as an appropriately safe environment to make social connections and to combat isolation. The attempt to create an illusion of safe anonymity allowed users to take certain calculated risks. Some participants disclosed names as well as nicknames and also discussed their geographical location. Participants could be e-mailed from the message thread as a web based e-mail connection appeared next to each post. In addition to disclosures made by participants, there were also many ways of identifying and tracking users built into the software.  

The sensibility of anonymity that participant's language conveyed was constructed on a spatial characterisation of the web site. Participants described their interaction in this forum as being removed from their immediate physical locality and constructed the forum as a 'place' away from where they lived. These descriptions positioned Gaygirls.com as 'utopic', in the sense of a mythical extrapolation or projection from lived realities, a space of fantasy communion that temporarily satisfies the desire to belong. In relation to research ethics, the construction of such 'utopic spaces' nested within commercial domains of the Internet, is a cause for concern, as it substantiates claims that participants are not fully aware of the conditions under which they communicate through Internet technologies. This point is reiterated in Frankel and Siang and the ‘AoIR ethics working committee - a preliminary report’ (Frankel & Siang, 1999). The latter report details some of the discussions of the AoIR ethics working committee members. This committee was, in part, established to respond to the earlier, Frankel and Siang, report. The AoIR report notes in 'Section 4, Provisional Results, B' that an important aspect of research ethics is the information available to users and in particular that the "ethically-relevant facts of using IT" (AoIR, 2001) need to be part of both user and researcher education. However, the fact that some participants on Gaygirls.com made statements indicating that they would be harmed if their physical and online identities were linked, does not necessarily mean that they were unaware of the technical limits of the medium or miscalculated the levels of risk. Participants' statements of stratification and utopic construction strengthen their 'suspension of disbelief', strengthening the illusion of safe space. This strategy engenders trust and connectivity and creates community feeling. Ultimately, however, the utopic safe space that is created is not a material condition.  

Despite the use of spatial descriptors, the web site remains very much a textual production. The sense of spatiality is constructed through linguistic techniques such as use of metaphor and simile. Use of language in the forum was open, there was, for example no use of exclusive textual codes. There were very few short-hand acronyms such as reported in other forums (Wakeford, 1996), and those which were deployed were transparent, such as 'u' for you or 'lol' for laugh out loud. The text was decipherable to the inexperienced CMC user and to heterosexual participants and thus not rarefied. CMC-specific codes were rare but oblique lesbian specific codes even more so. The web site functioned to invoke inclusion on the pre-existing terms of a lesbian hierarchy, and by setting up a strategic coalition of inclusion the term 'lesbian' was reworked as inclusive. However, inclusion often involved a reification of lesbian hierarchies of authenticity. This strategy of using potentially inclusive, but actually specific terms, maps onto an individualistic logic where economic considerations see a catering to niche markets in the face of the fragmentation of consumer groups. Online communities simultaneously commodify identity and utopian desire, working through a liberal agenda of identity without exclusion, to maximise market share whilst capitalising on fragmentation. 

The Misplaced Calculation of Risk

The formalities of print publication are well established and it is difficult to get into print accidentally or unknowingly. In contrast, what the Frankel and Siang's report describes as the ‘ease of conductivity’ (Frankel and Siang, 1999, p. 9) also relates to the ease of dissemination of computer-mediated texts. As with any computer operation, it is easy to execute a command that does not have a series of checks and simply 'sends' or 'prints'. Gaygirls.com has a statement advising participants that all messages posted automatically become information in the public domain. This information is not directly connected to the forum analysed, however. The web site owners use language that positions Gaygirls.com overall as a highly visible text: there are press clippings and links to other sources, to on and offline media, and reviews of the web site. Activism, visibility and publicity are stressed and Gaygirls.com champions representation as a means of awareness raising, education and critiquing issues of sexuality and identity. The forum analysed is not separated off from the rest of the site through either rhetorical or technical strategies. Participants using the system are not for example advised that they are 'entering a chat room' or any other kind of domain that might evoke a perception of relative privacy. The language used to frame the forum is drawn from the conventions of public speaking resonant of the 'speakers corner' convention in the UK. 

The web site producers are realistic about the risks of disclosure and warn participants about this in the 'about' pages. They do not signpost or differentiate specific parts of the site however, and we feel that this is a central issue of concern. An aspect of the site that consists of a previously prepared interview or article will be 'read-only'. In contrast, the forum discussed in our research provides an opportunity to air an opinion in a similar way to the letters page in a print publication. The site producers do use the word 'public' at several points but this does not necessarily warn participants that these sections of the web site are equivalent in terms of publication to a letters page in a print publication such as a newspaper. 

The Frankel and Siang report (1999) calls for a delineation of public and private in relation to the Internet. We feel however, that it is more important to raise levels of awareness that technically there can be no such delineation. Although there is a shared ethical responsibility across the participants, the researchers and the site producers, the immediate responsibility for raising public awareness that material placed on the Internet is always open to scrutiny lies primarily with the web site producers who actively invite participants to contribute. In the long term though, the use of a 'suspension of disbelief' to construct community is not something a site producer should necessarily police and it is at policy level that this kind of information needs to made available. The fact that the producers of Gaygirls.com did not contribute to this awareness raising is arguably in part because they are engaged in a political awareness raising of LGBT visibility. It is also because the production of any textual forum in CMC occurs primarily through the use of particular textual signifiers. Therefore a warning about risks related to disclosure situated in the same text as the forum would contribute signifiers of risk to the language and potentially hinder the production of the quality of engagement described above.  

The Political Imperative for Visibility

The producers of Gaygirls.com had made a strategic choice to construct a supportive environment, without differentiating between different sections of the web site. This has a range of implications, but we should not simply conclude that the web site producers were either exploitative or irresponsible. A primary social condition of Western capitalism is that commodification enters into every aspect of life. In this kind of social context, strategically producing a supportive web site based on inclusion and political visibility for a potentially marginalised group could be seen to outweigh the imperative to protect individual participants who fail to become fully aware of the terms of engagement.  

Gaygirls.com represents an attempt to create political visibility for a group by creating a centre for activity, networking and discussion. The politics of representation create an imperative for the researcher to preserve the same level of visibility. LGBT identities and communities are underrepresented in traditional print and broadcast media. The ease of communication available through Internet technologies gives marginalized or subcultural groups access to a medium of statement that has the potential to challenge this underrepresentation. Academic discussion of subcultural groups can potentially add to their cultural capital, legitimise and increase acceptance of the diversity of culture, challenging the monolithic and dominant conceptualisation of society as structured through the heterosexual matrix (Butler,1993).

When we first wrote up our research we decided to describe the research process but to withhold the name of the web site and participants so as to prevent connecting a user to their text in a context that they had not chosen. The object of our research was to contribute to knowledge of the coming-out narrative in relation to lesbian identity and to examine how coming out online related to contemporary ideas of the lesbian community. This way of writing up met the aims of our research. When we came to publish the research however we were asked, through peer review, to supply user names in order to make the research more engaging to the reader and to give the article more 'colour'. We particulary didn't want to disclose user names because they are often not anonymous but pseudonymous (Turkle, 1996, and Danet, 2001). They are traceable to users and cannot be separated from offline names. When thinking through these issues, the statement of one participant — that she would experience harm if her communication around this web site and her family life were linked — was a primary factor. Although most of the participants were open about engagement and had no issues with disclosing geographical location and photographs, many of these users were under the age of legal consent. These factors were the final influence on our decision to use pseudonyms.

Clearly, we could not bring attention to the activity around this web site in a way that would link users or user names to texts. To do so would be to remove participants' choice and control over their own stratification of audience. However, this sense of control is illusory, and the fact that participants are engaging in activities without an awareness of the conditions of participation is problematic. The site producers had indicated that, in general terms, messages posted to the site were public information. The problem as we see it is how a greater and wider understanding of the conditions of Internet use can be developed. This issue is one that has been raised by other researchers but although the emphasis placed on educating the research community is important (Frankel & Siang, 1999) we feel that this emphasis should also be directed toward promoting education about the conditions of Internet use as part of the compulsory education curriculum. 

In writing up the research we agreed to use pseudonyms to make the research more readable and we created names with a careful consideration of how these reflected the signifiers chosen by users. Within the framework of political representation, however, this is a misappropriation. We felt that this was not a technique that we would recommend and not one that we would use in future but it was the best model under the conditions through which we were working. During the timescale that we carried out the research we felt that we were piecing together elements from different models, none of which were entirely satisfactory. We felt that there was an irresolvable tension between models that positioned the material as text and those that treated it as person.  

Our original decision to change user names and disguise Gaygirls.com might be interpreted as contributing to the homophobic myth that LGBT identity is something to be ashamed of, is pathological or is dangerous. From this perspective, there is an argument that our use of the conventions of the human subjects research model was actually unethical. As in the example of Indymedia already discussed, the decision to disguise online activity, justified through the a rhetoric of 'protection' may result in furthering the unequal power relations of media production by blocking full representation of alternative media. 

Conclusion

Spatial descriptions of the Internet, prevalent in advertising and political jargon, are reinforced in academic research that deploys the human subjects model. Yet only some types of Internet use can usefully be described in spatial terms. Academic research that positions the Internet as a social space containing cultural activity ripe for observation ignores the range of textual applications that the Internet supports. Evidence of harm to individuals resulting from some Internet research indicates that in some instances application of a code of research ethics akin to the human subjects model is appropriate. This harm does not justify uncritical application of such a model to research of any material drawn from the Internet, however. As Internet researchers, we need to develop models that acknowledge both spatial and textual understandings of the Internet. Such models are required if we are to produce research that examines the complex intersection of technologies, form, genre and content that the Internet supports. 

The Web version of the Internet is a relatively recent cultural phenomenon, and patterns of its use are still emergent. As indicated in other research, the use of Internet technologies emerged from a pattern of usage that involved specialists (Abbate, 1999). An example is the ARPANET researchers who both used and helped define the use of communication networks. The release of Internet technologies in the mass market, post 1994, popularisation via the web, and the advertising rhetoric that can be summed up as 'it's whatever you want it to be' contribute to a situation where users are engaging with a process with which they are partially unfamiliar. Users may be less informed about the issues involved in textual production via the Internet than in print media or traditional broadcast media such as the radio or television. In societies where the Internet is promoted through politically and commercially motivated rhetoric of self-determination, there is a need for education and government policy that addresses the risks of Internet use (Hamelink, 2000). This notwithstanding, researchers cannot afford to assume that individuals engaging in cultural practices through the Internet are always at more risk than those using other media. Rather than unintentionally rarefy the Internet through the promotion of research models that cloak it in spatial metaphors, we might attempt to contribute to the debate about its nature.

There are issues and rights at stake in these debates other than those of privacy and safety. The Internet user is also entitled to a degree of representation and publication in the public domain. If an individual or group has chosen to use Internet media to publish their opinions then the researcher needs to consider their decision to the same degree that they would with a similar publication in traditional print media. Overly protective research ethics risk diminishing the cultural capital of those engaging in cultural production through Internet technologies, and inadvertently contributing to their further marginalization. For members of LGBT communities, visibility, and the rights to self-represent and to self determine are political imperatives in a dominant mediascape that continues to marginalise and pathologise them. To maintain a research model akin to the human subjects model would be to risk impeding and potentially eliminating promising research. It is not always possible, for example, to gain the consent of a large number of participants who may have changed their email address or ceased posting to a web site on which the material under research is located. This should not prevent research of textual material that they have chosen to output via the Internet anymore than it would for textual products in other print or broadcast media. Academic research ideally endeavours to reflect the range and versatility of the media that it considers and we should avoid any unintentional erasure of minority groups in research that might result from considering their textual output as private social interaction.  

The debate cannot be reduced to a matter of deciding between a spatial or textual characterisation, or between the human subjects model and the paradigm of publication. Instead we should attempt to find a way of acknowledging the hybridity of the Internet, acknowledging that the texts it supports are neither virtual selves nor objects completely distinct from those who write them. We have advocated here a return to consideration of issues debated in academic disciplines that analyse hybrid texts, such as English Literature, Life Writing, and other feminist research methodologies. We draw specifically on the feminist oral historian, Salazar's, notion of the 'relations of production' (Salazar, 1991, p. 98) of a text. A reading of the text that includes the content, form, history and researchers' response all need to be considered as part of these relations of production and power is a central issue.  

These relations include, for example, statements of intention in the text about whether it is intended for publication or for consumption by a circumscribed group of people. This needs to be evaluated as part of the meaning of the communication, regardless of the legal status of such utterances. This applies to statements by users as well as those written by web site producers. Other questions arise over issues such as warnings to users that material is publicly accessible, the temporal relationship between production and dissemination and the genre of the web site. These are evaluative issues, but these relations will effect how users perceive their acts of participation and production: for example, is the site like a public forum, modeled on a graffiti wall or a message board, or is it more like a newsletter intended only for members of a group? These structures clearly contribute to the user's perceptions of their activity on it. Features such as a log -in process or the layering of a message board 'behind' other pages may also contribute to the user's perception of privacy and safety.  

Content is a controversial issue, although Herring (1996a) makes a case that it could be considered irrelevant in certain circumstances. A question that has pervaded research debates is that of special provision if the subject matter of the text pertains to certain subcultural groups. An assumption that sensitive material should be accorded special privacy rights should not be made unconditionally, as the initial definition of 'sensitive' is ideological and is itself a process that is subject to ethical questions. The history of the British press, for example, shows that social power is played out and contested through the dissemination, alteration and suppression of texts (Curran & Jean Seaton, 1997). This history, (very generally) of alternative and independent production becoming jeopardized and partially suppressed because of the political economy of the mass media, need not automatically be repeated in relation to the Internet.  

We feel that the commonly deployed spatial models of the Internet, and the concomitant conflation of the body with data, are issues that deserve further enquiry. These models appear to be seeing a resurgence through the dominant construction of the Internet as a place in which human subjects act. We argue that a hybrid model of relational ethics that incorporates text, space and bodies needs to be applied to Internet research. The human subjects model only has limited application, and where it is applied indiscriminately or as the 'safest' option it in fact can be unethical and not 'safe' at all. 

 

Acknowledgments

The writers are indebted to Dr Sally R. Munt for her co-authorship of the publication referred to in the case study (Munt, Basset & O'Riordan, 2002) and to Professor Charles Ess for his support and editorial work. 

 

References

Janet Abbate. Inventing the Internet. MIT Press, Cambridge and London, 1999.

M. H. Abrams. A Glossary of Literary Terms. Holt, Rinehart & Winston, New York, 1941.

AoIR (Association of Internet Researchers). AoIR ethics working committee - a preliminary report. 2001. Retrieved June 25, 2002 from http://aoir.org/reports/ethics.html

Roland Barthes. Death of the Author. In Image Music Text. Fontana Press, London , 1977. (Trans Stephen Heath).

Caroline Bassett. The Arc and the Machine Graduate Research Seminars. In Media and Cultural Studies Series. University of Sussex, 23rd October, 2001.

Nancy K. Baym. Tune In, Log On: Soaps, Fandom, And Online Community. Sage Publications, London , 2000.

David Bell and Barbara Kennedy, editors. The Cybercultures Reader. Routledge, London and New York, 1999.

Michael Benedict. Cyberspace: First Steps. MIT Press, Cambridge, 1993.

Katherine Borland. That's not What I Said': Interpretive Conflict in Oral Narrative Research. In S. B. Gluck & D. Patai, editors, Women's Words: The Feminist Practice of Oral History. Routledge, New York and London, 1991.

Amy Bruckman. Community Support for Constructionist Learning. Computer Supported Cooperative Work, 7:47-86, 1998.

Vannavar Bush. As We May Think. Atlantic Monthly, July, pp. 47-61, 1945.

Judith P. Butler. Bodies that Matter: On the Discursive Limits of "Sex". Routledge, London, 1993.

Edward A. Cavazos. Cyberspace and The Law: Your Rights and Duties in the On-Line World. MIT Press, Cambridge, 1994.

Susan Clerc. Estrogen Brigades and 'Big Tits' Threads: Media Fandom Online and Off. In Cherny and Wiese, editors, Wired Women: Gender and New Realities in Cyberspace. Seal Press, Washington, 1996.

Copyright, Design and Patents Act 1988 (c.48). Retrieved April 18, 2002, from http://www.hmso.gov.uk/acts/acts1988/Ukpga_19880048_en_3.htm(mdiv16

James Curran and Jean Seaton. Power without Responsibility: The Press and Broadcasting in Britain. Routledge, London, 1997.

Brenda Danet. Cyberpl@Y:Communicating Online. Berg, Oxford, 2001.

T. Dowling. Chatroom of the Week. The Guardian, G2, p.23. April 18, 2002.

Terry Eagleton. Literary Theory: An Introduction. Blackwell, Oxford, 1983.

Michel Foucault. The Archaeology of Knowledge. Tavistock, London, 1972.

Mark S. Frankel and Sanyin Siang. Ethical and Legal Aspects of Human Subjects Research on the Internet: A Report of a Workshop June 10-11, 1999. November, 1999. Retrieved April 1, 2002 from http://www.aaas.org/spp/dspp/sfrl/projects/intres/report.pdf

S. B. Gluck and D.Patai. Introduction. In S. B. Gluck and D. Patai, editors, Women's Words: The Feminist Practice of Oral History. Routledge, New York and London, 1991.

Stine Gotved. Newsgroup Interaction as Urban Life. AoIR: State of the Interdiscipline, Conference Paper. University of Kansas, September 14-18, 2000.

Cees J. Hamelink. The Ethics of Cyberspace. Sage, London, 2000.

Donna Haraway. Primate Visions. Verso, London, 1992.

Sandra Harding. Whose Science? Whose Knowledge? Cornell University Press, Ithica, 1991.

Katherine N. Hayles. How We Became Posthuman: Virtual Bodies in Cybernetics, Literature and Informatics. Chicago University Press, Chicago, 1999.

Susan C. Herring. Linguistic and Critical Analysis of Computer-Mediated Communication: Some Ethical and Scholarly Considerations. The Information Society Vol. 12, No. 2: 153-168, 1996.

Susan C. Herring, editor. Computer-Mediated Communication: Linguistic, Social and Cross-Cultural Perspectives. John Benjamins, Amsterdam, 1996.

Matthew Hills. Virtually Out There: Strategies, tactics and Affective Spaces in On-line Fandom. In S. R. Munt, editor, Technospaces: Inside the New Media. Continuum, London, 2001.

Christine Hine. Virtual Ethnography. Sage, London, 2000.

M. Jolly. Editor's Note. In Encyclopedia of Life Writing: Autobiographical and Biographical Forms. Fitzroy Dearborn Publishers, London and Chicago, 2001.

Robert Jones. The Ethics of Research in Cyberspace. Internet Research, Vol. 4 No. 3, 30-35, 1994.

Fred Kaplan. Dickens: A Biography. Hodder and Stoughton, London, 1988.

Lori Kendall. MUDer? I Hardly Know 'Er: Adventures of a Feminist MUDer. In Cherny and Wiese, editors, Wired Women: Gender and New Realities in Cyberspace. Seal Press, Washington, 1996.

Storm King. Researching Internet Communities: Proposed Ethical Guidelines for the Reporting of Results. The Information Society, Vol. 12, No. 2, 1996.

George Landow. Hypertext 2.0: the Convergence of Contemporary Critical Theory and Technology. John Hopkins Press, London and Baltimore, 1997.

J. Lye. The 'Death of the Author' as an Instance of Theory. 1996. Retrieved April 24, 2002 from http://www.brocku.ca/english/courses/4F70/author.html

G. E. Marcus and M. Fischer. Anthropology as Cultural Critique: An Experimental Moment in the Human Sciences. University of Chicago Press, Chicago, 1986.

Annette Markham. Life Online: Researching Real Experience in Virtual Space. Altamira Press, London, 1998.

Vicki Mayer. When the Camera Won't Focus: Tensions in Media Ethnography. Feminist Media Studies, Vol. 1, No. 3, 2001.

Hugh Miller and Jill Arnold. Breaking Away from Grounded Identity? Women Academics on the Web. CyberPsychology & Behavior, Volume 4, 1:95-108, 2001.

Laura Miller. Women and Children First: Gender and the Settling of the Electronic Frontier. In Brooks and Boal, editors, Resisting the Virtual Life. City Light Books, San Francisco, 1995.

Hans Moravec. Rise of the Robots. Scientific American, December, 124-135, 1999.

Sally R. Munt. Heroic Desire: Lesbian Identity and Cultural Space. Cassell, London, 1998.

Sally R. Munt, editor. Technospaces: Inside the New Media. Continuum, London, 2001.

Sally R. Munt, E. H. Basset and Kate O'Riordan. Virtually Belonging: Risk, Connectivity and Coming Out On-line. The International Journal of Sexuality and Gender Studies. Volume 7, No. 2, 2002.

Robert Pepperell. The Post-Human Condition. Intellect, Oxford, 1995.

Howard Rheingold. The Virtual Community: Homesteading on the Electronic Frontier. Addison-Wesley, Reading, 1993.

Paul Ricoeur. Hermeneutics and Human Sciences: Essays on Language, Action, and Interpretation. J. Thompson, editor. Cambridge University Press, Cambridge, 1982.

C. Salazar. A Third World Women's Text: Between the Politics of Criticism and Cultural Politics. In S. B. Gluck and D. Patai, editors, Women's Words: The Feminist Practice of Oral History. Routledge, New York and London, 1991.

Theresa M. Senft. Homecam Heroines: Gender, Celebrity and Performance on the World Wide Web. Peter Lang Publishing, New York. Forthcoming: Spring 2002.

S. Seidman, editor. Jurgen Habermass on Society and Politics: A Reader. Beacon Press. Boston, 1989.

Barbara Sharf. Beyond Netiquette: The ethics of Doing Naturalistic Discourse Research on the Internet. In Steve Jones, editor, Doing Internet Research: Critical Issues and Methods for Examining the Net. Sage, London, 1999.

Beverley Skeggs, editor. Feminist Cultural Theory: Process and Production. Manchester University Press, Manchester, 1995.

Liz Stanley, editor. Feminist Praxis: Research Theory and Epistemology in Feminist Sociology. Routledge, London and New York, 1990.

Richard Stallman. The GNU Operating System and the Free Software Movement. In Chris Dibona, Sam Ockman and Mark Stone, editors, Open Sources: Voices from the Open Source Revolution. O'Reilly, London, 1999.

Allucquere Rosanne Stone. The War of Desire and Technology at the Close of the Mechanical Age. MIT Press, Cambridge, 1995.

Malin Sveningsson. Creating a Sense of Community: Experiences from a Swedish Web Chat. In Linkoping: Linkoping Studies in Art and Science. 2001.

T. Minh-Ha Trinh. When the Moon Waxes Red: Representation, Gender, and Cultural Politics. Routledge, New York, 1991.

Tiziana Terranova. Post-Human Unbounded: Artificial Evolution and High-Tech Subcultures. In Bell and Kennedy, editors, The Cybercultures Reader. Routledge, London, 2000.

David Tetzlaf. Yo-Ho-Ho and a Server of Warez: Internet Software Piracy and the New Global Information Economy. In Andrew Herman and Thomas Swiss, editors, The World Wide Web and Contemporary Cultural Theory. Routledge, London and New York, 2000.

Sherry Turkle. Life on the Screen: Identity in the Age of the Internet. Weidenfeld & Nicolson, London , 1996.

Nina S. Wakeford. Sexualised Bodies in Cyberspace. In W. Chernaik and S. Deegan, editors, Beyond the Book: Theory, Text and the Politics of Cyberspace. University of London, London, 1996.

A. Ward. Copyright and Oral History, Oral History Society. Retrieved April 5, 2002, from http://www.oralhistory.org.uk/ethics

Kevin Warwick. Cybernetic Organisms: Our Future? Proceedings IEEE, Vol. 87, No. 2, 387-389, 1999.

Margaret Wertheim. The Pearly Gates of Cyberspace: A History of Space from Dante to the Internet. Virago, London, 1999.

William Wimsatt and Munroe Beardsley. The Intentional Fallacy. In David Richter, editor, The Critical Tradition: Classic Texts and Contemporary Trends. Bedford, Boston, 1954.

Endnotes

1 This is not to say that we should revert to a model that views all Internet texts as equivalent to print media (e.g. Cavazos, 1994). BACK

2 Dickens took issue with the publication of his works in the USA, from which he did not receive an income (Fred Kaplan. Dickens: A Biography. Hodder and Stoughton, London, 1988). BACK


Introduction
Charles Ess

Ethical Issues of Online Communication Research
Rafael Capurro & Christoph Pingel

What is special about the ethical issues in online research?
Dag Elgesem

Research Ethics in Internet-Enabled Research: Human Subjects Issues and Methodological Myopia
Joseph B. Walther

Studying the Amateur Artist: A Perspective on Disguising Data Collected in Human Subjects Research on the Internet
Amy Bruckman

Representations or People?
Michele White