Lisa Nathan, Batya Friedman
"What does policy have to do with me?" That was the essence of the question Jonathan Lazar put on the table in his inaugural article for interactions. We agree with him that the answer is "a lot." Our recent efforts to work with the United Nations International Criminal Tribunal for Rwanda, officials from the Rwandan government, Rwandan NGOs, institutional review boards, and our own diverse project team continue to be an intense lesson in how different levels of policy exert influence on the design of interactive information systems .
Earlier articles have discussed large-scale public policies that can influence design decisions on a national or even international level (e.g., Web-accessibility legislation and voting-machine standards). In this article we consider the difficulties that arise when these large-scale policies are in tension with smaller-scale policies (e.g., institutional, organizational, or even project-based). We suspect these tensions are becoming more prevalent as HCI researchers and designers undertake increasingly complex projects in domains such as sustainability, health, government, and justice. The interactive systems that result from these endeavors can cross numerous boundaries among legislation, regulations, principles, standards, and norms (i.e., written and unwritten policies). These boundaries are often amorphous, and, more often than not, more than one level of policy is at play in any given set of design decisions. For purposes of this discussion, we conceptualize the term "policy" quite broadly and find it to be closely linked with social values. Pragmatic in structure, policies help HCI researchers and designers to identify and choose among alternative courses of action according to a specified set of goals or values.
First, a quick look at five levels of policy that HCI researchers and designers may encounter in their work:
Domestic (government) policy. Instituted by national, state/provincial, and perhaps city governing bodies, these are the principles, rules, regulations, and laws that typically come to mind when considering public policy. The Access to Justice Technology Principles developed in the state of Washington provide one useful example . Written to guide court administration, these principles articulate how courts can use information technology to give all Washingtonians fair access to the state's justice system.
International policy. In some arenas, national governments have come together to create treaties, pacts, and even international laws that transcend national boundaries. For example, the data-protection principles and policies established by the European Commission govern the flow of personal information among the 27 EU member states and three EEA (European Economic Area) member countries (Norway, Liechtenstein, and Iceland) with third-party countries .
Organizational/institutional policy. Though not public policy per se, the organizations and institutions with which HCI researchers and designers work to develop interactive systems may have their own policies that impact system design. Many organizations, for example, have explicit policies that provide guidelines for "appropriate" employee Internet and email usage.
Working-group norms and expectations. Again, though not public policy and perhaps not even explicitly stated, the specific groups of individuals who will be using the technology and perhaps even participating as co-designers may have tacit but strongly held norms and expectations relevant to a technology under development. These norms and expectations may function as if they were explicitly agreed upon policies; to the extent they do, they may need to be accounted for in the design process and eventual system deployment .
HCI researcher and designer principles, policies, norms, and expectations. Finally, HCI research and design teams have their own explicit or tacit commitments similar to the working-group norms and expectations that can function as if they were policies, guiding acceptable design practices and outcomes. For example, project team policies might guide the breadth and depth of user studies, the minimal extent to which co-design must occur, and explicitly supported values for a given project. Seen in this light, Bidwell and Winschiers-Theophilus advocate project team policies that encourage Western researchers and designers to engage in co-design with Africans within an African context .
To help explicate the ways in which various levels of policy can interact within a single research and design project, we draw from one of our current projects. Our work with video interviews from the United Nations International Criminal Tribunal for Rwanda (UN-ICTR) begins where the official tribunal records end: the collection of 49 in-depth video interviews with a diverse group of tribunal personnelprosecutors, defense counsel, judges, investigators, interpreters, court administrators, and otherson the experience of participating in the tribunal. We refer to this project as the "Voices from the Rwanda Tribunal" (or "Tribunal Voices") . The original goals were to provide widespread public access to the interviews for access and reuse while protecting against revisionist histories.
To enhance the "Tribunal Voices" material on a local area network, we explored the development of a commenting system. The goal of the online commenting system was to provide visitors to information centers (managed by United Nations employees, many from Rwanda) in Kigali and roughly a dozen Rwandan provinces with the opportunity to contribute their reflections and analyses about specific clips and the tribunal in general. The ability to weigh in might scaffold open discourse in a country reaching for democracy, in turn supporting UN policies of international human rights and development . In the words of one of our African colleagues: "This website can help people to reconciliation.... You want people to talk. To get rid of the hatred inside them." Yet in early discussions about the specifics of the commenting system, strong concerns arose. The information-center employees worried that some comments might be perceived as violating Rwanda's 2008 Genocide Ideology Law (e.g., comments that deny the genocide). Violators of this law face harsh penalties (fines and/or incarceration), and the site would likely be blocked in Rwanda.
This concern for national policy led our colleagues to urge for a "moderated" forum in which all posts would be reviewed prior to posting. If the online forum on the local area network were moderated and anti-government messages removed, would this system be supporting open, democratic discourse? Together with our co-designers we decided to set up a moderated forum for the near term; we plan to revisit this choice as the political climate in Rwanda continues to evolve.
Conducting research in a post-conflict situation poses unique challenges for protecting human subjects. In particular, prior to being in the field, it is difficult if not impossible to understand communication norms and the conditions that would be perceived as safe by potential participants. During our first attempt to recruit UN-ICTR personnel to participate in our project, our Institutional Review Board (IRB) required that we extend at most two invitations to each potential participant. We soon learned that this "at most two invites" policy contradicted tacit understandings within the UN-ICTR culture. As one participant explained: "If you really want me to participate, you need to tug on my arm, so I know that you are truly interested in what I have to say." The actions determined by the two-invites policy were perceived in the UN-ICTR culture as polite politicinterest that was likely feigned. We realized we were losing tribunal personnel who wanted to participate by following a policy created in the American context to protect participants working in an international culture. Upon returning to our home university, we discussed with the IRB the implications of overly prescriptive recruitment policies when working internationally. Swayed by our experience, the IRB has allowed more adaptable policies for our future work in Rwanda.
In contrast to the two situations here that concerned interactions with policies from outside the project team, our last example highlights a policy tension that arose from within. Alerted by prior work to the usefulness of articulating explicitly supported project values , early on the project team established a list of eight "Guiding Principles," one of which was the principle of access. While presumably in agreement, the team discovered in the thick of making design decisions that members had different understandings about what constitutes meaningful access and for whom.
In particular, the team debated whether to release the video interviews in the languages in which they had been conducted (primarily English) or to wait until translations and subtitles in the native language for Rwanda (Kinyarwanda) could be obtained, which could take up to two years. At stake was who would be able to access the video interviews and whennot only technically but also linguistically. Some team members strongly favored releasing the videos as soon as possible, without Kinyarwandan subtitles, prioritizing access to the international legal community for important, time-critical work. Other team members were just as much in favor of waiting to release videos until a Kinyarwandan translation was available, interpreting "supporting access" in a robust sense to include the broader Rwandan populace. We delayed making a decision until a follow-up trip to Rwanda enabled us to engage diverse Rwandan communities in the decision-making process. The response from the Rwandans we spoke with was surprisingly unified: It was better to post in English sooner rather than wait for translation into Kinyarwanda. At the same time, many underscored the importance of eventual translation to enable Rwandans in rural areas greater access to the material. Key to this discussion: While the project team had articulated guiding principles, the team had not adequately defined those principles or put in place a process for setting priorities when tensions inevitably would arise.
In the midst of designing a particular system, it can be difficult to hold on to the reality that policy conditions will shift. As our experiences working on "Tribunal Voices" highlight, governments will change, policies will be reworked, and a project team's ability to work with policies will evolve. Thus, a further challenge for HCI designers and researchers entails working with and building multi-lifespan information systems that can be responsive to and take advantage of those shifts when they do occur .
This article has explored policy tensions through a politically sensitive international case study. We expect interaction designers and researchers in other contexts to face analogous challenges as different levels of policy bump up against each other. In response to the policy tensions discussed above, we remain faithful to our original goals of widespread public access and reuse of the "Tribunal Voices" materials, albeit within the confines of existing public-security and domestic law. We welcome insight on how other researchers and designers have approached similar challenges.
We thank the participants from the United Nations International Criminal Tribunal for Rwanda for sharing their experiences, reflections, stories, and insights. We also thank the many Rwandan people and organizations that worked with us, including Never Again Rwanda, Hope After Rape, and the ICTR Documentation and Information Centres. The members of the "Voices from the Rwanda Tribunal" team contributed to all aspects of this work. This material is based in part upon work supported by the National Science Foundation under Grant Nos. 0325035 and 0849270, UW Foundation, and generous gifts to the "Tribunal Voices" project. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of our donors.
1. For more information, please see http://www.tribunalvoices.org/
2. Access to Justice Technology Principles; http://atjweb.org/frontpage/
3. European Commission Justice and Home Affairs-Data Protection; http://ec.europa.eu/jus-tice_home/fsj/privacy/index_en.htm/
6. Friedman, B., Nathan, L. P., Grey, N. C., Lake, M., Nilsen, T., Utter, E., Utter, R. F., Ring, M., and Kahn, Z. "Multi-lifespan Information System Design in Post-conflict Societies: An Evolving Project in Rwanda." Extended Abstracts of CHI 2010 (2010): 28332842.
7. United Nations' Office of the High Commission for Human Rights; http://www.ohchr.org/EN/Pages/WelcomePage.aspx/
8. Borning, A., Friedman, B., Davis, J., and Lin, P. "Informing Public Deliberation: Value Sensitive Design of Indicators for a Large-scale Urban Simulation." Proceedings of ECSCW 2005 (2005): 449468.
Lisa P. Nathan is an assistant professor at the School of Library, Archival and Information Studies, University of British Columbia. She teaches courses in information policy, information ethics, and a survey course on the field of information studies. In addition to ongoing work with the Voices from Rwanda project, Nathan is investigating agile methods to support both information system designers and policy makers in efforts to improve the influence of information systems on human values, particularly those of ethical import.
Batya Friedman is a professor in the Information School at the University of Washington where she directs the Value Sensitive Design Research Lab. Known for her pioneering work in value sensitive design, Friedman is currently working on methods for envisioning and multi-lifespan information system designnew ideas for leveraging information systems to shape our future. These adaptive solutions fundamentally co-evolve technology, policy and social structure. The "Voices from the Rwanda Tribunal" is an initial project in this multi-lifespan information system design effort.
Figure. The ambitious "Voices from the Rwanda Tribunal" project includes a collection of 49 video interviews with tribunal personnel. The videos are viewable at htp://www.tribunalvoices.org/videos.html. All are subtitled in Kinyarwanda.
©2010 ACM 1072-5220/10/0900 $10.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2010 ACM, Inc.