In Conversation with Elizabeth Churchill
I got a chance to catch up with Elizabeth Churchill, PhD. and Principal Research Scientist with Yahoo! Research to chat about some key topics from this month’s DUX2007 in Chicago (you can still register). The following is an excerpt of our conversation.
Elizabeth, fair warning—the first couple of questions are going to be a bit like the “Justify Your Existence” column in The Onion. They are intended as a warm-up. Describe what you do in 5 words or less?
Understand people to design experiences
Wow, good answer. Why is it worth seeing you at DUX?
Because I’m fascinating, nosey, curious, and interested in what you all do.
Ok, well the theme of DUX2007 asks some pretty fundamental questions related to the notion of design and the designer. How do you perceive the roles of design and designer changing?
My background isn’t design with a capital “D” but I like designing experiences, and developing ways for people to create their own experiences. It’s interesting working for an internet company because I am at the heart of how that is really changing – more and more, people have access to tools and malleable content, and to other people.
In the past, people principally knew others through proximity and so this is what we supported in our research and design. I designed for learning groups, for workplaces and workgroups, and for social groups made up of people who largely knew each other or were part of the same organizational cultures or social groups, even if they had never met face to face before. Now at Yahoo!, I’ve shifted to thinking about experiences that connect people who may never meet in person, who come from different worlds, and who know nothing of each other, but can do amazing things together. How can we build tools for them to co-create and accomplish outstanding outcomes together?
Speaking generally, in the past the social structures and behaviors were predictable because mostly people knew each because they were lived close together, or they were related or they traveled and met other people in the same jobs and class. We could make certain assumptions based shared knowledge, shared etiquettes, social mores and overlapping social networks. Of course there are always exceptions, people who traveled, explored moved through different groups and social places. But on the Internet, we can stay on our sofas, embedded in many social worlds at once, and we can’t necessarily predict who we will interact with, and what cultures and societies will overlap simultaneously.
All models of trust and intimacy, trust and kinship, signals of respect and mutual understanding are being tested. People are doing a lot more work in trying to understand the complexities of how these work online, but we still haven’t figured it out.
Speaking of that, we both have academic training in the social sciences, but when we were in school the tools and methods were created for different contexts. I did a short study on World of Warcraft and attempted to use traditional observational methods to understand group dynamics. I found the traditional tools difficult and possibly inadequate? What’s your perspective on this as a researcher?
Well, I think differentiation between the tool and the use of that tool practice is important. The tool of observation, for example, it’s not just being seen and seeing. It’s bigger than that. Real observational understanding in the social sciences, as stressed in anthropological training in particular, involves getting in there – considered, longer-term participation, being immersed in the role(s), having conversations, conducting interviews, and studying artifacts in their production and use - anything to gain multiple perspectives and a deep understanding of that world. Depending on the design intent, we need to develop more or less “thick” descriptions of the worlds we are designing for – but it is important to first be clear about our design intent, and to use the methods that are most suited.
So, I don’t think the tools are the problem but instead tools are being used in a shoddier fashion. If you get really deeply into an online situation with traditional tools, you have a very good chance to understand. One person who does this really well is T.L. Taylor in her book where she describes her work studying online games. Necessarily, she spends a lot of time within gaming worlds and uses tools and methods of observation in a really considered way.
I don’t feel like I’m forced to develop new tools but I would really like to find a way to work on this at Yahoo! – to have an impact on data collection that blows away boundaries between qualitative and quantitative camps. We both know the division between these forms of data doesn’t really exist; you can make anything qualitative quantitative and deciding what to collect in a log is a qualitative decision. Of course,. In any situation where we are collecting log data, privacy would have to be a primary concern. I also really want to address how what happens online affects the offline and, as a result, develop a deeper understanding of how the internet fits into everyday life in the myriad ways it does.
Too often though, in our field, the qualitative work to understand experiences people are having and from which we can design is not being done effectively. I would like to see us bring it to another level. I do think new tools are needed, which is really the subtext of your question, but not necessarily for the thinking part of analysis – that takes patience, training and perseverance – you can’t understand complex practices and the implications for a design problem in 5 minutes. Rather, as far as tools go, it’s about collection, storage and access. Observational data can be hard to manage and observation isn’t enough if we don’t have access to collected rich data and notes to mull over later; online that means images, videos, text, audio and so on. And activity logs are usually not intended for analysis later. It is amazing to me that designers and developers don’t routinely think about those logs as they are building the tools. So we end up with something that is a machine dump that needs massive post hoc scraping and cleaning before we can understand what is going on. Managing, indexing, sorting, selecting and conceptually and practically integrating all of these forms of data is a huge challenge. As a whole, we need to have better log hygiene, in terms of what we collect, how we collect and what we do as a result.
In part, we need new tools because the artifacts and data objects we are using in analysis are more collectable than ever before. The tools need to develop specifically to support this collection. But we also need to be smarter about how we define research questions. As I’ve said before to my students, “You can do the thinking before or after data collection. But you can’t escape the thinking part.” The nature of ethnographic methods mean we try not to be driven in our observations by strong hypotheses which may blind us to what is actually going on, but that doesn’t mean we shouldn’t understand what area we wish to understand and be careful about what data we collect. The alternative requires more sorting and possibly more research later. It is all too tempting to record everything, and then feel overwhelmed and never look at the collected data again.
Switching gears, it seems at some level like people are closer than ever before because of social networking platforms but also further apart than ever: alienated. What do you have to say to that?
I think there needs to be a couple of levels of meaning and reaction to this assertion. First, it’s easy to confuse connection with communication. I don’t think the problem from a human perspective is any worse or better than it has always has been, but the scale – and thus the potential for us observing and experiencing this connection-communication mismatch is really high. Right now the potential to find people to connect with is nearly limitless. I can meet people who are really far away and develop a connection with them but how that connection will or will not develop beyond connection and into a relationship (or not) is still really complex. The technologies also uncover mismatches between people who do know each other – one can be in a relationship where “pinging” someone constantly on IM or SMS makes you feel intimate, close, connected and loved, but can make the other person feel alienated and assaulted. I know people who have had to negotiate the annoyance and disappointment of this kind of mismatch, and who have had to learn together to manage their connectivity before their relationship could deepen.
Some of the studies we’re doing at Yahoo! make it clear that human decision-making around the development of intimacy hasn’t really changed. We see the fundamental negotiation of comfort with other individuals in nearly every interaction. Some of the technologies which are becoming more popular around how you negotiate relationships and intimacy are really complex and their popularity isn’t unrelated. But, being a friend is not about clicking a button to say “I have now friended you”. We always have to negotiate, to find each other’s comfort zones around connections and sharing, and what’s really interesting is that these technologies are making a lot of those decisions explicit.
Because of the many more people we interact with, we also have to set really explicit boundaries. I’ve had to be much more gauche recently online than I would have been in the past and in person – tacitly saying “No you are not my friend”, when in the past I may have subtly walked away at a party, now I am having to take an action – or not take an action. We underestimate the power of body language to really communicate our feelings about others, and how the subtlety of the cues we send are understood – the message gets through, but we don’t have to actually offend someone. Often not thinking of someone as a “friend” is not about their capacity to be a friend or not, it is about the fact that you just don’t know yet. In these online spaces, I have to say “Are you my friend or are you not my friend?” The “No” comes as a rebuff that could be read as “absolutely not and you will never be”, when it may just mean “I don’t know yet, I only just met you.” This is explicit and is a fascinating part of one of the studies I’ll be presenting at DUX.
There are some ways that technologies can enhance experience with those whom you are already intimate. For example, sending a picture from a cell phone to a person who you are already intimate with can bridge physical distance a little through emotional touch. But as I said, these kinds of touches need to be matched between people, and often a sign of intimacy is when the rhythm of reaching out to each other becomes something we understand, we know, but we don’t have to think about. Think about a situation when you knew someone so well that, without planning it, you knew when they were going to be in touch. We develop a shared rhythm, and go from a discordant, jerky dance of communication to a well executed foxtrot of dialogue – and it is that temporal matching that is a big part of intimacy. (I choose the foxtrot because I remember having to do all kinds of fancy footwork when turning corners, while maintaining a smooth balance and rhythm of the dance.) These different technologies are allowing us to express potential and actual intimacies in the ways we do (and don’t) develop rhythms of communication across various new means and modes. To me, the temporally synched touches signal intimacy as much, if not more, than the content of the “messages”.
So we happened to meet at a past conference and then flickr helped to develop our friendship. We ended up having some pretty intimate interactions based on some photos we shared and I feel like we became closer. At the same time, it was clear we hadn’t really intended for it to happen—it just sort of did. Because of that experience and a few creepy experiences with followers of my blog, I ended up writing a piece about “Personal Brand” and how you really have to control the online face of yourself like we never had to before. What’s your opinion on this?
I think we’ve always done impression management. I have seen it most explicitly in Japan and in the UK. But conversational or communicational impression management in the past has been about a lot about face to face behavior management or highly structured non-proximate communications like letters. What we now have are texts of ourselves which we share with people and then become public – all my pictures, and my web page, and so on. Of course, many people manage others impressions of them through their clothes, houses, cars and so on. But these tend to be relatively well understood. The problem is that we don’t comprehend the full ramifications of these texts online, An example is when something has been cached, people come across it and don’t understand it is like someone going through an adolescent journal and thinking that the content deeply reflects the 30 year old who is standing in front of them – as they are now. That is such a dumb error, but people are prone to it. On the internet the oldest of pages still can look bright and shiny and new – and current. But also, even current things are stripped of rich content, they do not tell you the story of the person behind them, but they can leave that person vulnerable to misunderstanding - and intentional or unintentional assault.
One of the fundamental points about humanity is that we are very bad at imagining outside of our experiences. Imagined audiences are a very important concept. We say, “No one will be looking at me and what I do,” and this forms a protective shield. It’s security through obscurity. This is not a model we need to go forward with because it just isn’t the case at all.
I have hope that as designers we will create better experiences to see who is looking. We have to educate through clearly communicating impact; the other is educating through stories about what happens to people; and then there is a cultural education where people move en masse to first consider then protect systems or groups. This does not need to be conservative to be very effective. Domestic violence is a good example. Before efforts to make this problem visible, people didn’t believe it was happening. Once it was clearly communicated, and stories were told, society and culture can act to protect -- I should say we still have a long way to go in that regard, but at least there is recognition and action. Back to the internet, when we start to recognize and collectively address some of the injustices and problems around internet content and connectivity that hurt people as well as the wonderful benefits we get, we can start to design not just technologies but policies, standards and social mechanisms to address some of the problems we are seeing. As a society of internet users, we are in the early days, and things need to develop, to evolve. One can see this happening in discussions of governance in many online worlds like WOW.
On an individual level, we are just starting to understand that people are our there looking at us.
If there was exactly one concept or meme to be aware of in the social networking space what would it be?
Trust. Re-negotiating trust. How you and I understand trust is going to have to change in the world of Internet systems. There are many ways in which the boundaries breakdown and are still to be defined. We’ve talked about people, trust and intimacy. But what about service providers, what about companies like Yahoo! Do I trust the “system”, iPhone, Apple, the service, company, the federal government, do I trust the person who is looking at me on Flickr? Economically, politically, who do I trust? Who is going to take the hit when something bad happens? We as designers have a responsibility to ethically design around issues of trust—to inform the user of what will be done with information they share, who will have access to it, what they can and can’t change, that they can rely on a service that hosts their carefully crafted content. You need to be clear what you are doing with their information and communicate it before, during, and after interactions. It’s not just a personal matter, not just something we have to consider from a single user perspective. It is a business matter too; if consumers lose trust in a business, the business will pay.
Trust is the big one.
As always, thanks for taking the time to chat.
No problem at all. I’ll see you at DUX.