CSCW 2016 (ACM’s conference on Computer Supported Cooperative Work and Social Computing) took place in San Francisco last month. I attended (my second time at this conference!), and it was wonderful meeting new and old colleagues alike. I thought I would share some reflections and highlights that I’ve had from this year’s proceedings.
Many papers addressed issues of privacy from a number of perspectives. Bo Zhang and Heng Xu study how behavioral nudges can shift behavior toward more privacy-conscious actions, rather than merely providing greater information transparency and hoping users will make better decisions. A nudge showing users how often an app accesses phone permissions made users feel creepy, while a nudge showing other users’ behaviors reduced users’ privacy concerns and elevated their comfort. I think there may be value in studying the emotional experience of privacy (such as creepiness), in addition to traditional measurements of disclosure and comfort. To me, the paper suggests a further ethical question about the use of paternalistic measures in privacy. Given that nudges could affect users’ behaviors both positively and negatively toward an app, how should we make ethical decisions when designing nudges into systems?
Looking at the role of anonymity, Ruogu Kang, Dabbish, and Sutton conducted interviews with users of anonymous smartphone apps, focusing on Whisper and YikYak. They found that users mostly use these apps to post personal disclosures and do so for social reasons: social validation from other users, making short-term connections (on or off-line), sharing information, or avoiding social risk and context collapse. Anonymity and a lack of social boundaries allowed participants to feel alright venting certain complaints or opinions that they wouldn’t feel comfortable disclosing on non-anonymous social media.
Investigating privacy and data, Peter Tolmie, Crabtree, Rodden, Colley, and Luger discuss the need for articulation work in order to make fine-grained sensing data legible, challenging the notion that the more data home sensing systems collect, the more that can learned about the individuals in the home. Instead, they find that in isolation, these personal data are quite opaque. Someone is needed to explain the data and provide contextual insights, and social and moral understandings of data – for subjects, the data does not just show events but also insight into what should (or should not) be happening in the home. One potential side effect about home data collection might be that the data surfaces practices that otherwise might not be seen (such as activities in a child’s room), which might create concerns about accountability and surveillance within a home.
Related to privacy and data, Janet Vertesi, Kaye, Jarosewski, Khovansakaya, and Song frame data management practices in the context of moral economy. I found this a welcome perspective to online privacy issues, adding to well-researched perspectives of information disclosure, context, and behavioral economics. Using mapping techniques with interviewees, the authors focused on participants’ narratives over their practices, finding a strong moral undertone to the way people managed their personal data – managing what they considered “their data” in a “good” or “appropriate” way. Participants spoke about managing overlapping systems, devices, and networks, but also managing multiple human relationships – both with other individuals and with the companies making the products and services. I found two points particularly compelling: First the description that interviewees did not describe sharing data as a performance to an audience, but rather a moral action (e.g. being a “good daughter” means sharing and protecting data in particular ways). Second, that given the importance participants placed on the moral aspects of data management, many feared that changes in companies’ products or interfaces would make it harder to manage data in “the right way,” rather than a fear of inadvertent data disclosure.
I was happy to see continued attention at CSCW to issues of policy. Casey Fiesler, Lampe, and Bruckman investigate the terms of service for copyright of online content. Drawing some parallels to research on privacy policies, they find that users are often unaware of sites’ copyright policies (which are often not very readable), but do care about content ownership. They note that different websites use very different licensing terms, and that some users may have a decent intuition about some rights. However, there are some licensing terms across sites that users often do not expect or know about – such as the right for sites to modify users’ content. These misalignments could be potentially problematic. Their work suggests that clearer copyright terms of service could be beneficial. While this approach has been heavily researched in the privacy space to varying degrees of success, there is a clear set of rights associated with copyright which (at the outset at least) would seem to indicate plain language descriptions may be useful and helpful to users.
In another discussion of policy, Alissa Centivany uses the case of HathiTrust (a repository of digital content from research libraries in partnership with Google Books) to frame policy not just as a regulating force, but as a source of embedded generativity – that policy can also open up new spaces and possibilities (and foreclose others). Policy can open and close technical and social possibilities similar to the way design choices can. Specifically, she cites the importance of a specific clause in the 2004 agreement between the University of Michigan and Google that allowed the University the right to use its digital copies “in cooperation with partner research libraries,” which eventually led to the creation of HathiTrust. HathiTrust represents an emergent system out of the conditions of possibility enabled by the policy. It’s important to also recognize that policies can also foreclose other possibilities – for example, the restriction to other library consortia excludes the Internet Archive from HathiTrust. In the end, Cenitvany posits that policy is a potential non-technical solution to help bridge the sociotechnical gap.
I was similarly pleased to see several papers using the lens of infrastructure. Ingrid Erickson and Jarrahi investigate knowledge workers’ experience of seams and creating workarounds. They find both technological and contextual constraints that create seams in infrastructure – such as public Wi-Fi access that doesn’t accommodate higher bandwidth applications like Skype, incompatibility between platforms, or contextual examples might include locations of available 4G and Wi-Fi access or cafes that set time limits on how long patrons can use free Wi-Fi. Workers respond with a number of “workarounds” when encountering work systems and infrastructures that do not fully meet their needs: bridging these gaps, assembling new infrastructural solutions, or circumventing regulations.
Susann Wagenknecht and Matthias Korn look at hacking as a way to critically engage and (re)make infrastructures, noting that hacking is one way to make tacit conventions visible. They follow a group of German phone hackers who “open” the GSM mobile phone system (a system much more closed and controlled by proprietary interests than the internet) by hacking phones and creating alternative GSM networks. Through reverse engineering, re-implementing parts of the system, and running their own versions of the system, the hackers appropriate knowledge about the GSM system: how it functions; how to repair, build, and maintain it; and how to control where and who it is used. These hacking actions can be considered “infrastructring” as they render network components visible and open to experimentation, as well as contributing toward a sociotechnical imaginary foreseeing GSM as a more transparent and open system.
Adding to a growing body of CSCW work on time, Nan-Chen Chen, Poon, Ramakrishnan, and Aragon investigate the role of time and temporal rhythms in a high performance computing center at a National Lab, following in the vein of other authors’ work on temporal rhythms that I thoroughly enjoy (Mazmanian, Erickson & Harmon, Lindley, Sharma, and Steinhardt & Jackson). They draw on collective notions of time over individual ones, finding frictions between human and computer patterns of time and human and human patterns of time. For instance scientists doing research and writing code have to weigh the (computer) time patterns related to code efficiency and (human) time patterns related to project schedules or learning additional programming skills. Or they may have to weigh the (human) time it takes to debug their own code versus the (human) time it takes to invest time in getting another person to help debug their code and make it more efficient. In this timesharing environment, scientists have to juggle multiple temporal rhythms and temporal uncertainties caused by system updates, queue waiting time, human prep work, or other unexpected delays.
Cooperation and Work
Contributing to the heart of CSCW, several research papers studied problems at the forefront of cooperation and work. Carman Neustaedter, Venolia, Procyk, and Hawkins reported on a study of remote telepresence robots (“Beams”) at the ACM Ubicomp and ISWC conferences. Notably, remote tele-presence bot use at academic conferences differed greatly from office contexts. Issues of autonomy are particularly interesting: is it alright for someone to physically move the robot? How could Beam users benefit from feedback of their microphone volume, peripheral cameras, or other ways to show social cues? Some remote users used strategies to manage privacy in their home environment by blocking the camera or turning off the microphone, but had a harder time managing privacy in the public conference environment, such as speaking more loudly than intended. Some participants also created strong ties with their remote Beams, feeling that they were in the “wrong body” when their feed transferred between robots. It was fascinating to read this paper and compare it to my personal experience after seeing Beams used and interacting with some of them at CSCW 2016.
Tawanna Dilahunt, Ng, Fiesta, and Wang et al research how MOOCs support (or don’t support) employability, with a focus on low socioeconomic status learners, a population that is not well understood in this environment. They note that while MOOCs (Massive Open Online Courses) help provide human capital (learners attain new skills like programming), they lack support for increasing social capital, helping form career identity, and personal adaptability. Many low socioeconomic status learners said that they were not able to afford formal higher education (due to financial cost, time, family obligations, and other reasons). Most felt that MOOCs would be beneficial to employment, and unlike a broader population of respondents, largely were not concerned about the lack of accreditation for MOOCs. They did, however, discuss other barriers, such as lack of overall technical literacy, or how MOOCs can’t substitute for actual experience.
Noopur Raval and Dourish bring in concepts from feminist political economy to look at the experience of ridesharing crowd labor. They bring in notions of immaterial labor, affective labor, and temporal cultural politics – all of which are traditionally not considered as “work.” They find that Uber and Lyft drivers must engage in types of affective and immaterial labor, needing to perform the identity of a 5-star driver, pressured by the frustration that many passengers don’t understand how the rating systems are weighed. Their status as contractors provides individual opportunities for microbranding, but also creates individual risks that may pose customers’ desires and the drivers’ safety against each other. Through their paper, the authors suggest that we may need to reconceptualize ideas of labor if previously informal activities are now considered work, and that we may be able to draw more strongly from labor relation studies and labor theory.
Research Methods and Ethics
Several papers provided reflections on doing research in CSCW. Jessica Vitak, Shilton, and Ashktorab present survey results on researchers’ ethical beliefs and practices when using online data sets. Online research poses challenges to the ways Institutional Review Boards have traditionally interpreted the Belmont Report’s principles of respect, beneficence, and justice. One finding was that researchers believe that ethical standards, norms, and practices differ relative to different disciplines and different work fields (such as academia or industry). (I’ve often heard this discussed anecdotally by people working in the privacy space). However, the authors find that ethical attitudes do not significantly vary across disciplinary boundaries and in fact there is general agreement across five practices which may serve as a set of foundational ethical research practices. This opens the possibility for researchers across disciplines, and across academia and industry, to united around and learn from a common set of data research practices.
Daniela Rosner, Kawas, Li, Tilly and Sung provide great insight into design workshops as a research method, particularly when things don’t go quite as the researcher intends. They look at design workshops as sites of study, research instruments, and as a process that invites researchers to reflexively examine research practices and methods. They suggest that workshops can engage with different types of temporal relations – both the long lasting and meaningful relationships participants might have with objects before and after the workshops, and the temporal rhythms of the workshops themselves. What counts as participation? The timing of the workshop may not match the time that participants want to or can spend. Alternate (sometimes unintended or challenging) practices brought by participants to the workshops can be useful too. They provide important insights that might make us rethink about how we define participation in CSCW (and perhaps in interventionist and participatory work more broadly), and how we can gain insights from interventional and exploratory research approaches.
In all, I’m excited by the directions CSCW research is heading in, and I’m very much looking forward to CSCW 2017!