The field is made, not found: what presence buys and what it owes

The field site is not a place you find and then visit. It is a construct assembled from a research question, a set of connections and a decision about where to stand. Once assembled, it makes claims on the researcher. Access has to be renegotiated. Relationships shape what people do while they are watched. Consent does not survive the calendar. This third of ten units argues that constructing the field, not the length of the visit, is the method's first commitment. Long presence is a design decision with a price. The seven units that follow take up what the field yields.

The site is a construct before it is a place

Geertz (1973) made interpretation the object of the method. Culture, he argued, is not a power that causes events but a context within which they can be intelligibly described. Ethnographic description is interpretive: it works on the flow of social discourse and rescues what was said into terms a reader can examine.

His own account of entering a field is the raid. He and his wife reached a Balinese village in April 1958 and were ignored for the first days, until armed police broke up a cockfight they were watching. They fled with the crowd and hid in a courtyard. By the next morning the village had begun to talk to them. Geertz calls this the turning point of his fieldwork and is clear that it is not a recipe: a vice raid is no generalisable route to rapport.

Marcus (1995) extended the object beyond one location, arguing that the classic single-site study is inadequate to phenomena distributed across a world system. He set out tracking strategies that follow the people, the thing, the metaphor or the plot. He named the anxieties the shift produces, including the attenuation of fieldwork's power.

Blomberg and Karasti (2013), reviewing 25 years of ethnography in computer-supported cooperative work, put the construct at the centre. They note that the multi-sited programme has itself been criticised for paying too little attention to the bounding, selection and choice that constitute a site.

The online field was first bracketed as a place, then dissolved into everyday life

Hine (2000) established the internet as a field by refusing to decide in advance what kind of place it is. She treats it as both a site for cultural formations and a cultural artefact shaped by people's understandings. She is candid that the distinction replays the real and the virtual, in a way that can obscure how that boundary is itself constructed.

Kozinets (2002) gave the online field a procedure: netnography, ethnography adapted to online communities, illustrated on an online coffee newsgroup. He specified entree, collection, analysis and ethics. He presents netnography as faster, simpler and less expensive than traditional ethnography. It also selects its participants: a person with no reliable line, or on a metered data plan, leaves no trace for the ethnographer to find.

Hine (2015) then dissolved the split she had earlier bracketed. The internet, she argues, is embedded, embodied and everyday rather than a separate domain, so the practical questions become where to go, what to do there and how to acquire robust knowledge.

Pink and colleagues (2016) turn that position into principles, among them multiplicity, openness, reflexivity and a refusal to let digital media sit at the centre of the design. They organise the principles around seven concepts that run from experience and practice to localities and events.

Duration buys an account of a setting that a survey cannot reach

The clearest evidence for duration is a study that used it. Nardi (2010) reports more than three years of participatory research in World of Warcraft play and culture in the United States and China, including a month in Chinese internet cafes. A survey of players could not have produced her account of how players collaborate, learn and form friendships: a population of millions is not the same achievement as an understanding of a setting.

Boellstorff and colleagues (2012) codify the practice for virtual worlds. Their handbook centres participant observation, draws on work in EverQuest, Second Life, There.com and World of Warcraft, and takes the researcher from choosing a fieldsite to writing and publishing, with attention to review boards and consent.

Staying produces obligations that belong in the design

Duration is not free, and Blomberg and Karasti (2013) are the honest source on the bill. In their account access is a continuing negotiation rather than a permission granted once. The researcher's presence belongs in the data, because people act differently while they are watched. The horizons of inquiry keep expanding without settling what counts as proper ethnographic research.

Boellstorff and colleagues (2012) make the consequences concrete. Consent, review board approval and the handling of logs, images and video are part of the design of a virtual-world study rather than an administrative appendix. Kozinets (2002) adds the reflexive requirement, that the researcher who joins an online community must account for their own participation in what they then report. The obligation scales with the field's construction rather than with the time spent.

For HCI the import's hardest question is what the account is for

Hughes and colleagues (1994) reflected on fieldwork across several projects and domains rather than a single control room. They describe how they used ethnography to inform design. One problem was the difficulty of turning an account of work into something a designer can act on.

Their worked project is air traffic control. Four weeks of fieldwork in the London Air Traffic Control Centre were followed by a debriefing with the designers and then by a prototype. That cycle ran about four times before the team judged that more fieldwork would add little. What the ethnography supplied was not a list of requirements but an account of how a team of controllers stays jointly aware of the traffic, in part by reading, annotating and passing the paper flight strips.

Dourish (2006) states the objection. If the major evaluative criterion for an ethnographic study is the implications it provides for design, the metric may fail to capture the value of the investigation and may distort the practice. His claim is that this is the wrong standard, not merely a demanding one.

Button and Dourish (1996) give the tension its sharpest form. A method whose findings describe how members organise a setting cannot be converted into prescriptions without the description ceasing to describe the setting as it was, so the warrant for a design claim has to be produced rather than assumed.

A field can be an algorithmic system or a movement between settings

Seaver (2017) takes the constructed field to its hardest case. He argues that algorithms should be approached as multiples enacted through the varied practices of the people who engage with them. They should be studied as culture rather than as closed technical objects. His tactics deliberately do not depend on identifying a single algorithm or on achieving access. They work from the partial and mobile position of an outsider.

Gariglio (2024) reaches a comparable conclusion from the other direction. He reflects autoethnographically on fieldwork in documentary photography and in academic ethnography. He uses the hybrid field to argue that the field is produced through the researcher's movement between settings and modes, and writes the reflection partly as a methodological novel.

Contested question: what counts as a field, and what counts as its yield?

Two disagreements run through this literature. The first concerns the field's shape. Marcus (1995) holds that a single bounded site is inadequate to the phenomena. Nardi (2010) and Boellstorff and colleagues (2012) show what sustained presence in one setting produces. Blomberg and Karasti (2013) record the criticism that multi-sited work neglects the bounding and selection that make any site a site. Seaver (2017) pushes furthest, arguing that the outsider's partial position can do work that access cannot.

The second concerns the yield. Hughes and colleagues (1994) treat design relevance as the point of the exercise and report its practical difficulty. Dourish (2006) and Button and Dourish (1996) hold that the demand for design implications is the wrong measure and damages the method, while Pink and colleagues (2016) sidestep the dispute by grounding quality in principles rather than in a deliverable.

What would settle either question is not another position but a comparison. One setting would be studied twice, once through bounded single-site immersion and once through multi-sited or outsider tactics, with the accounts tested against the same recorded events. A second comparison would trace design decisions back to the fieldwork said to have informed them, which the literature describes as rare.

The two sides do not carry equal evidence. Marcus (1995) argues from theory, while Nardi (2010) reports three years of presence in one setting and Blomberg and Karasti (2013) review 25 years of practice. The design-yield dispute has one worked case, the air traffic control cycle (Hughes et al., 1994), against two critiques (Dourish, 2006; Button and Dourish, 1996). Seaver (2017) offers one study of outsider tactics. While the dispute is open, a designer should record how the site was bounded, which connections were followed, and what the consent covers. The strongest objection is that documenting the construction shows the route without showing that it reached anything a longer study would not.

Where the argument is heading

The direction of travel is away from measuring fieldwork by time spent and towards making the construction explicit. Hine (2015) relocated the difficulty to where to go rather than to presence as such. Seaver (2017) argues the tactics work from where the researcher stands. Gariglio (2024) treats movement between settings and modes as the field itself.

The sources this unit rests on

• Blomberg, J. and Karasti, H. (2013). Reflections on 25 years of ethnography in CSCW. Computer Supported Cooperative Work (CSCW), 22(4-6), 373-423. doi:10.1007/s10606-012-9183-1 Reviews 25 years of ethnography in CSCW and treats the field site as a construct extending beyond the workplace. Records that the difficulty of drawing design implications produced strategies for coupling ethnography and design.
• Boellstorff, T., Nardi, B., Pearce, C. and Taylor, T. L. (2012). Ethnography and Virtual Worlds: A Handbook of Method. Princeton: Princeton University Press. A handbook centred on participant observation in virtual worlds including Second Life and World of Warcraft, taking the researcher from choosing a fieldsite to publishing. Consent and review procedures are treated as design decisions. https://press.princeton.edu/books/paperback/9780691149516/ethnography-and-virtual-worlds
• Button, G. and Dourish, P. (1996). Technomethodology: paradoxes and possibilities. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 19-26. doi:10.1145/238386.238394 Names the paradox in using ethnomethodologically informed ethnography to build systems: the method is analytic and describes how a setting is organised rather than what it should become.
• Dourish, P. (2006). Implications for design. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 541-550. doi:10.1145/1124772.1124855 Argues that treating implications for design as the main evaluative criterion for ethnographic work in HCI may fail to capture what such studies are worth. The standard, not the study, is at fault.
• Gariglio, L. (2024). On fieldwork in the hybrid field: a methodological novel on ethnography, photography, fiction and creative writing. Qualitative Research, 24(2), 444-456. doi:10.1177/14687941221149584 An autoethnographic reflection on fieldwork in documentary photography and in academic ethnography. Argues that the hybrid field is produced through the researcher's movement between settings and modes, so the field is made rather than found.
• Geertz, C. (1973). The Interpretation of Cultures. New York: Basic Books. Develops thick description and a semiotic concept of culture: culture is a context within which events can be described rather than a power that causes them. The unit's foundational account of ethnography as interpretation. https://www.basicbooks.com/titles/clifford-geertz/the-interpretation-of-cultures/9780465097197/
• Hine, C. (2000). Virtual Ethnography. London: Sage. doi:10.4135/9780857020277 Establishes the internet as both a site for cultural formations and a cultural artefact shaped by people's understandings, and notes that the distinction replays the real and the virtual. Asks what an ethnography taking both seriously would look like.
• Hine, C. (2015). Ethnography for the Internet: Embedded, Embodied and Everyday. London: Bloomsbury. Argues that the internet is embedded in everyday life rather than a separate domain, so the questions become where to go, what to do there and how to know it. Moves this unit past the online and offline split. https://www.bloomsbury.com/uk/ethnography-for-the-internet-9780857855701/
• Hughes, J., King, V., Rodden, T. and Andersen, H. (1994). Moving out from the control room: ethnography in system design. Proceedings of CSCW '94, 429-439. doi:10.1145/192844.193065 Reflects on using ethnography across several projects and domains rather than one setting, and sets out ways fieldwork was used to inform design with the benefits and problems of each.
• Kozinets, R. V. (2002). The field behind the screen: using netnography for marketing research in online communities. Journal of Marketing Research, 39(1), 61-72. doi:10.1509/jmkr.39.1.61.18935 Sets out netnography as ethnography adapted to online communities, with procedures for entree, collection, analysis and ethics, illustrated on an online coffee newsgroup. Presents it as faster and less expensive than traditional ethnography, the counterpoint this unit has to answer.
• Marcus, G. E. (1995). Ethnography in/of the world system: the emergence of multi-sited ethnography. Annual Review of Anthropology, 24, 95-117. doi:10.1146/annurev.an.24.100195.000523 Argues that the classical single-site study is inadequate to objects distributed across a world system, and sets out tracking strategies that follow people, things, metaphors and plots. Names the anxieties the shift produces.
• Nardi, B. (2010). My Life as a Night Elf Priest: An Anthropological Account of World of Warcraft. Ann Arbor: University of Michigan Press. doi:10.3998/toi.8008655.0001.001 Reports more than three years of participatory research in Warcraft play and culture in the United States and China, including a month in Chinese internet cafes. A worked case of what duration produces that a survey cannot.
• Pink, S., Horst, H., Postill, J., Hjorth, L., Lewis, T. and Tacchi, J. (2016). Digital Ethnography: Principles and Practice. London: Sage. Sets out digital ethnography as principles, including multiplicity, openness and reflexivity, organised around seven concepts from experience to events. Grounds the claim that practice can drive the field's construction. https://archive.org/details/digitalethnograp0000pink
• Seaver, N. (2017). Algorithms as culture: some tactics for the ethnography of algorithmic systems. Big Data and Society, 4(2). doi:10.1177/2053951717738104 Argues that algorithms should be studied as multiples enacted in practice rather than as closed technical objects, with tactics that work from the partial position of an outsider without requiring access. The unit's sharpest challenge to the claim that long presence buys understanding.