Questions for Wired Magazine

From Cyborg Anthropology
Jump to: navigation, search

With Tom Cheshire

1. You mentioned machines trawling data, such as search items across sites. Are there other types of data too? How exactly do you analyse those, and the search items, then synthesize them into your research?

There are many other kinds of data that can’t be reached from a program or bot. These are usually in the form of stories or experiences given by groups of people.

As a cyborg anthropologist, my research is preformed on two field sites - the analog site of the analog world, and the digital world. The analog world has it’s own challenges. You can’t see into people’s minds, you have to be present to see their actions, and you often physically have to be present if you want to see what’s actually happening in a system of people. One of the ways to get around this is to identify a group to study and then identify certain key people in the group. In a stable group there’s usually a leader, in fluctuating groups there are sometimes multiple people vying for leadership. Group leaders store a great deal of data about the group, especially the core group members. Rapid field research necessitates quickly finding out who in the group is the group ‘historian’ so to speak. They’re a person who internally collects information about the group members. Often they are a fan, organizer, or someone heavily involved in the group. They understand, more than most others, how the group works. In anthropology, this person is called the informant. They’re invaluable to the anthropologist because they are willing to help the anthro understand how a culture or group works. Identifying the informant or informants is where I focus the most attention when I’m first entering a group.

The digital field site is the other place where I spend a great deal of my time. Unlike an analog site, my field site is often 15 inches wide and thousands of miles deep. The computer is my interface into a field-site the likes of which no traditional anthropologist has seen before. Now that people are offering up bits of their lives -- reporting them --- the field site is much richer - but it also presents other challenges. For instance, to find the stories and where they live online, one has to know how they speak, what they care about and how they relate to one another. Sometimes a small thread from a search leads to a whole tribe of people talking on a forum or series of blog sites. Once I know where a group of people is participating online, I can figure out greater details of the group. Does the group care about this product over the other? What if X or Y were changed? How would the group react? What makes the group happy, and what makes them mad? Who are the core participants of the group and how much power do they have other other group members? Who are the leaders, fans and followers? What about the outliers? How do new members enter the group? What interests or experiences lead people to find and join the group?

I try to find all of the places where a group of people gathers. If it’s people who love Toyota trucks or philosophy books, I try to find where they live online. Often, they don’t all live in one place, but in multiple places. Once I’ve found some of the places they live I begin to rank the places in order of size and importance. Then I can begin to dig in and see patterns among the community members. I try to find types of people who have similar desires and actions. I’m really looking for the elements that form systems. Once you understand a system (i.e. if A happens, B will happen) then you can understand how something will evolve over time. If you get enough points on a curve, you can predict behavior. This is often something I do when I create company strategy - seeing not what the consumer wants now, but what will be important to then in the future, and what technologies will be emerging or available at that time - how a consumer would consume or create in that system. What it all comes down to is minimizing risk to the user of changing their ritual. Providing a small change in their life that increases over time, or a change that reduces the amount of time and space it takes to do a previous process, then it will improve someone’s life. That’s a good goal to have.

I often find a dozen types of people, but understanding and figuring out the types of people that make up a group is not an easy task. I generally read thousands of entries and group them into sets of similar stories or responses. From there I can understand the cultural system as a whole. This generally has great use value to a company that wants to make a new product but doesn’t understand what their community is looking for. Because so many people are talking about their hopes, desires and stress online, it’s possible to query that field-site to see what would actually benefit them. Will the community hate or like this new change? How much would they pay for it? Would it help their lives over time, or will they be inhibited by it? How can their lives be helped by better designs? For companies selling traditional products, the answer is not usually on new social structures like Twitter - it’s elsewhere, and often on forums that have run for a decade.

I’ve been using NameGenWeb, a Facebook application by danah boyd and Bernie Hogan ( to graph Facebook graph data. I use UCINet and NetDraw to take the data from NameGenWeb and visualize it. (Sample group visualization: Something I’m interested in right now is how many major hubs of nodes people have on Facebook. Once I get permission from someone to download their social graph, I make an image an ask them to identify the clumps of groups. For instance, some say “those are my high school friends”, “I went to band camp with them”, and so on. I’m interested in how age affects the number of distinct social groups a person has on Facebook, and graphing social networks helps with easily understanding that data.

2. When you're conducting interviews, what questions do you ask?

I’ve seen best results when using an informal interview methodology. Generally, when you're doing research you want to ask questions that don't skew responses. You can’t, for instance, lead with something like “don’t you hate it when…” You might instead lead with an experience that leads the interviewee to relate one of their own. Experience matching is a normal and comfortable part of conversation that is used when anthropologists are doing participatory ethnography or “deep hanging out”. A good question or experience is often a seed that can open up a wider conversation about a topic to multiple people in a room. Instead of answering solid questions which often put people on the spot, conversation slowly develops, and the topic goes deeper than a simple interview can go. I like to ask questions that lead to stories. When you get a lot of the same stories from many different people, you know you're onto something important.

When I was researching Facebook, simply mentioning one's parents joining Facebook was enough to spur dozens of stories, sometimes leading into 3 hour discussions of awkward situations. From there it was easy to see the schism between how one experienced real life and the online world with their parents. In real life, teenagers talk in a certain way to other teenagers, adults talk in a certain way to other adults, and when adults and teenagers communicate, they have their own way of protecting their own boundaries of what they feel safe communicating. When parents began joining Facebook, they entered a territory that most teenagers were communicating amongst each other on. They were suddenly privy to the types of conversations teenagers had with each other and not adults. When those two groups kind of blended together in the online space, suddenly all of those boundaries got kind of muddy. That’s where the invasion of privacy and awkwardness appeared. It was the equivalent of someone’s mom hanging out with them while they were having casual conversation with their friends. Facebook didn’t have privacy settings or a system for extending these analog social boundaries online, so it was (and is) a really awkward time for many teenagers.

When I was doing research on mobile phones, I was trying to understand what made phones so compelling, and also the experience one felt when disconnected from the phone. I wanted to get stories about how people felt when their phone ran out of batteries. I often lead with my own experience of losing battery power. That would lead quickly to stories of similar events from those I was talking with. I quickly found that two experiences came up the most. One was of horrific loss of capability and a sort of disoriented powerlessness, as if a lifeline has beed cut off. Others reported slight surprise when their phones ran out of batteries, and then a sort of relaxed feeling afterwards. A feeling of living their life without checking their messages or thinking about who was trying to contact them.

Getting stories and grouping them into similar reactions was key to my research. If you get enough of these experiences and correlate them with things like age, gender and lifestyle, you can begin to see certain patterns in behavior and likelihood for one person to experience something over the other. For instance, one could crunch the numbers and be able to say that one’s income affects their likelihood of getting upset when their phone runs out of batteries and so on. This quantitative research can be done through entering coded qualitative data into a statistical analytics such as SPSS. ( Statistical analytic programs like SPSS are anthroplogist’s friends because they allow anthropologists to run correlations and other mathematical methods for finding out if one thing causes another thing to happen. My favorite is the Pearson's correlation, which allows one to find a correlation between at least two continuous variables. It’s a simple test, and if you have the data formatted correctly, can lend a lot of insight (like income level being an indicator of whether or not one will feel frustration over loss of battery power).

Sometimes there is no correlation, and sometimes there are unexpected correlations. Running correlative tests is like opening a present after a lot of hard work. You have your hypothesis of what you think the outcome might be, and sometimes that hypothesis is totally wrong and you run across something you never saw before. It’s very exciting when that happens, because it’s not the same old “well of course this affects that, everyone knows this already and doesn’t care”, but “no one has seen this before! It’s very fascinating”. (I can’t share the outcomes or work for confidential reasons).

For my current research on tech and startup communities I’m doing a longitudinal study. This is a study that will take years to complete. I’m 3 years in, and it will likely be 2 more years until I’ve seen enough to fully write research. This is because the community I am studying is changing over time, and I’m tracking the changes and events that lead to those changes. Economics, geography and politics have large effects on how the community has evolved, and these things happen on longer time scales than things such as memes, for instance. In this case, I take observations over a longer period of time.

3. How do you identify members of the tribe to speak to?

A lot of traditional anthropologists seek out the ‘other’, or a group that exists outside of their current society. Anthropologists may go to remote countries or study drug addicts or the wealthy, etc. What a number of researchers have realized is that the world around us is a perfect place to apply anthropology. There are a few difficulties in doing this. First, it’s harder to objectively observe and stay distant enough to make conclusions about systems and trends. On the other hand, it’s often easier because one might already understand the language and culture, so it takes less time to get accustomed to the culture. For instance, we live in a culture where everyone has these things in our pockets that cry, and we have to pick them up and soothe them back to sleep. But that’s not enough! We have to feed them every night by plugging them into the wall, and we have to upgrade them and take care of them so they don’t break. At no other time in history have we had this reality - these non-human devices that we take care of as if they are real, and we’re as dependent on them as they are on us.

Studying how people interact with each other through these little technosocial interactions, versus just the analog interactions, is something that fascinates me. I’m always in observation mode. One of the things I’m seeing more and more is people standing motionless in the middle of doorways, on top of steps and on sidewalks completely transfixed by their screens. When the majority of cell phone use was through voice, cell phones were more like cigarettes. If you used them inside, you’d pollute the space of others. If you took them outside, you’d have something to do while you stood around. And if you were bored and waiting for a bus you could make a call or two to pass the time. Now, smart-phones are being poked and prodded at vs. talked to. This dramatically changes the interaction and positioning of those using a device. It explains the motionless people in the middle of another experience quietly pausing themselves to a greater experience inside their devices.

In my research I’ve tried to make eye contact with those talking on a cell phone vs. those using a touchscreen. Whereas the talker is in an auditory space and mostly in the current moment, the quiet touchscreen user can be stared at and sometimes even talked to before breaking out of their task. My early research was mostly on talkers. Now it is transitioning towards touchscreen users. There is a definite difference in how the device is used, and new social issues are just beginning to arise because of it.

4. How do you do the observations? Again, as much detail as possible would be great.

Participant observation. For studies over time such as understanding developer communities and startup culture, I’ve actively participated in the culture, often organizing and contributing to it. Along the way, I’ve picked up terminology and habits that have taken me almost to the point of ‘going native’, which is kind of fun and kind of terrifying at the same time. The thing about technology is that it is exciting and addictive. It’s often difficult to stay objective and see it from a neutral perspective. Every so often I have to catch myself or apply what I’ve learned to software projects (apologies — this question is better answered through my response to question #2).

5. Are you planning to rejoin academia or is the wiki the way forward?

For now the wiki is a way to both make public my work and observations as well as provide a resource for those looking to study technology and culture. I’ve considered grad school (leaning towards MIT), but I was told to take time off between undergrad and grad school to see what the ‘real world’ was like before deciding what to really focus on in grad school. It will likely be a while before it becomes clear what really needs to be done in the field. In the end, cyborg anthropology is a placeholder term for an evolution of anthropological methods and study. It’s about using new tools to do fieldwork in new places, and to study all spaces and types of humanity, not just foreign ones. And when it comes down to it, the wiki is an extension of my brain. It’s a place to collect, store and build upon thoughts. It’s also an easy way to share them with others. For instance, if I get an E-mail about a particular theory or idea, and I have a page written about it in the wiki, I can link the page vs. write the response, because I’m really linking them to a part of my external brain. And because it is a wiki, they can add something to it if they feel it is incomplete. Wiki’s age well. They evolve and grow more complex and nuanced and useful over time. I’m only 6 months into this wiki. There’s still much more to be done.

Though I’m not officially affiliated with academia, I read academic papers constantly, especially those written 10-20 years ago about the coming “virtual reality” or those who talk about ubiquitous computing and other things. I’ve found that simply replacing “virtual reality” with social networks lends very well to those papers. In addition, I’ve been collecting sites from colleges and universities that contain writing about cyborgs and cyberculture. They’re rather hard to find today, because the idea of cyborgs and high tech reached its apex in the in the early 00’s and then kind of fizzled. Writing about the study ‘cyberculture’ and ‘virtual reality’ had a limited audience since the beginning of the information revolution. Some of the sites are still around, but they are quite obscure. Many are missing, which means that I rely on the Internet Archive as a research tool almost constantly. I’m always digging around the historical layers of the web. The wiki’s goal is to help resurrect some of those briefly-lived resources, understandings and predictions.

What I mean by predictions is that well-written papers about the coming world of virtual reality actually predict a lot about what we’re dealing with now. I feel like I’m doing a bit of future history when I do this research, as I’m often encountering worlds of people who have experienced what it was like to have an identity on the early Internet. What was experienced in labs 30 years ago by only a few people is now experienced by millions every day. There is a lot to learn from going back in time and reading the work of those who were there first. Many don’t realize that it’s already been done before and that those pioneers left stories, warnings and experiences on the way.

I read a lot of social theory. It gives me a lot of ground to stand on when I make my observations. Theodor Adorno, Marc Auge, Zygmunt Bauman, Marshall Berman, Walter Benjamin, C. Wright Mills, Michel Foucault and others like Guy Debord, Roland Barthes, Jean Baudrillard, David Harvey, Celeste Olalquiaga, Deleuze and Guattari, Paul Virilio. As far as more traditional anthropology goes, Guy Debord’s Society of the Spectacle and Erving Goffman’s Presentation of Self in Everyday Life are solid foundations for digital ethnography. Virilio writes about speed and culture, which is key to understanding the acceleration of changes that are occurring. Bruno Latour has Actor Network Theory that can help place humans and technology into a system that can be more easily studied.