What is the problem you're trying to solve?
Key to maintaining free expression on the Internet is understanding and responding to the hate, harassment and abuse that the communications network also delivers. Online harassment affects and inhibits use of many online platforms — including, notably, editing of Wikipedia — and can discourage new or continued participation or divert conversations away from constructive communication. While harassment affects people of all genders and backgrounds, online abuse often compounds historical discrimination against women and underrepresented groups and chills their participation in online forums where important issues are discussed and knowledge is shared.
Too often, solutions to the harassment problem are developed without reference to relevant research or understanding how that research might inform responses. While there is an understandable desire to build an app that will “fix” the problem, countering harassment and abuse requires interventions that leverage knowledge about the conditions, for example, under which bystanders are most likely to intervene on behalf of a victim, what interventions might work best to de-escalate attacks in different venues, and other insights new and existing research may provide.
There is little opportunity for researchers from different parts of the world to exchange research findings or to collaborate on new research on this topic. Because of cultural differences, similar research may produce somewhat different research findings, but there is no global forum to encourage discussion of research or collaborative work on countering harassment.
A further challenge to successfully reducing and responding to online harassment is that large Internet companies have limited incentives to share data about online harassment or proprietary research they commission. Collaborative work with academics may lead to an occasional publication, but the vast majority of the research insights and data that fuels them remain behind locked doors. To the extent that the analysis of large data sets, or interviews with perpetrators or victims, might provide insight into the problem and suggest interventions, researchers outside the big Internet companies often must start afresh without access to the most useful data.
What is your solution?
First, we want to build a research-driven approach to the development of counter-harassment interventions. We will examine the current state of research and identify the needs of different stakeholders, including users, community moderators and law enforcement.
There is a growing body of academic research focused on harassment in the online space, but there is also research on sexual harassment and abuse in offline settings that is often overlooked. Providing an easier way for individuals and entities designing interventions to identify, access and apply insights from the broad range of current academic research would raise the odds of success. Such a mapping would also identify gaps in the knowledge base and enable a more strategic approach to new research. We will hold a two-day research conference to assess needs and develop a research agenda that will be widely vetted and shared. We are aware that other academic institutions, for example the Berkman Center at Harvard, have held convenings on this topic. We intend to conduct an extensive review of research presentations and conference reports from prior events and to build on rather than replicate those efforts.We are also hoping that companies that have held much of their research closely, will share research insights that will inform new research and technical projects.
Second, we want to encourage specific research/technical projects that will contribute to solutions to address harassment by issuing a challenge to the University of California at Berkeley School of Information (I School) and the broader community to engage in research and the development of tools to address the harassment problem. The I School Challenge will reward a juried cash prize for the most important new research and/ or technical tools that are developed in response to the Challenge.
Third, we will build a research commons that will compile published research across disciplines and across borders, encourage researchers, technologists and others to contribute research findings and practitioners to post and discuss research questions and findings and develop new collaborations. Such a portal will also catalogue available tools, products, policies and other approaches so that anyone interested in working on the harassment issue can find valuable resources and a community of common purpose.
The Challenge aims to achieve three aligned objectives:
- build a research-based approach to countering online harassment;
- motivate development of counter-harassment research, tools and other projects; and,
- support collaboration through the sharing of research, data, policies, questions and tools.
Building a research-based approach
Hold a two-day research conference (January-March ‘17) to focus on needs assessment and a research agenda to inform solutions: we will examine the current state of research and identify the needs of different stakeholders, including users, communities and law enforcement.
In preparation for the conference:
- Students will develop extensive literature reviews to identify existing research and prepare background memos on these questions to provide a baseline of knowledge for the conference and help identify research gaps.
- Students will interview developers and users of multiple platforms (including Wikipedia) in order to: develop an understanding of the harassment challenges on the platform, the current tools and policies deployed by the platform, the affordances available to empower users to respond or otherwise address harassment; and to identify relevant research and gaps.
- We will also reach out through relevant Wikimedia venues to solicit input on the needs assessment and help identify relevant research and tools.
This pre-conference work will also help identify potential speakers and invitees. Conference papers will be prepared on the basis of this work to support the work of the conferees in identifying research needs, promising approaches and developing a research agenda.
After the conference:
- A conference report will be prepared, setting out key findings and proposing a course of action. We will widely share the draft of the conference report, first to solicit feedback and then to encourage broad participation in the execution of the plan.
Motivating the development of counter-harassment research, tools and projects
Organize a Challenge Prize in order to elevate the issue and incentivize participation across and beyond the Berkeley campus, and to recognize the most promising work in countering online harassment. Based on the model of other successful prize challenge programs, a prize provides both a financial and a reputational incentive to participation. Activities to support such a contest would include:
- Issuing a call to the I School community (students, faculty and alumni) as well as to the broader Berkeley community (e.g. Schools of Engineering, Psychology, Sociology, Journalism, Law) to participate in the Challenge by engaging in new research and development of tools and other approaches to afford Internet users a greater ability to address harassment. Interdisciplinary projects will be particularly encouraged.
- Hold regular campus events throughout the challenge year to discuss and critique projects with invited experts.
- Host a "Day of Code" (March-June 2017) to encourage interdisciplinary teams to develop technical tools to address harassment, drawing not only on technical skills, but on the assessment of needs and relevant research. (This event will build on the model of previous Social Innovation Hackathons held at the I School in 2013 and 2014.)
Supporting collaboration through the sharing of research, data, policies, questions and tools
Develop a portal for research and tools to compile published research across disciplines, encourage researchers and others to contribute research findings and practitioners to post and discuss research questions and findings and develop new collaborations. Such a portal (or "research commons") will also catalogue available tools, products, policies and other approaches so that anyone interested in working on the harassment issue would find valuable resources and a community of common purpose.
MediaWiki would be a natural software tool for a research repository of this kind. We might also use Discourse, an open source tool for managing community discussion that has specifically considered harassment and community issues in its design.
October 2016-2017, Total: $71,900
Project Management/Supervision: Sub-total: $30,000 (covers time of several people)
- Overall project management, internal/ external communications/ financial management
- Campus and community outreach
- Management/Supervision of background research documents and other material for conference
- Meeting with faculty from relevant fields to identify strains of literature and identifying students to participate in developing the research literature reviews for conference
- Issue the Community call to identify other relevant research
- Developing/outlining the plan for background research materials, supervising the students – individual/ group meetings, review of drafts, final editing, coordinating third party review
- Drafting overview document to accompany background material
- In consultation with others, develop conference design, selecting research questions for focus, identifying speakers
- Publicity, outreach to potential participants
- Manage logistics for conference: venue, travel for speakers, invitations, meals, AV etc (with student volunteers and administrative support)
- Conference management, including supervision/coordination of volunteers
- Summary Report and research plan. Assumes volunteer student note takers, first drafts and revisions that are developed throughout the conference through whiteboards/google docs/wikis and other means throughout conference which will then need to be turned into a summary, action and research plan, and revised based on community feedback.
- Management/supervision of the development of the Research Commons
- Outreach to interested students, faculty, community members to discuss needs, brainstorm design options etc.
- Ongoing supervision and support to student volunteers
- Ongoing outreach to global research community to identify and share research and technical tools
- Plan and execute communications plan launch event for the Research Commons
- Management of the Day of Code
- Event planning logistics
- Broad campus and community outreach
- Plan and execute communications plan
- “Day of” management including coordination of student volunteers
- Coordinating the I School Challenge
- Public Announcement/ I School event
- Ongoing promotion throughout I School and UCB: bi-monthly lunches with speakers, review/discussion of technical and research projects
- Coordinate selection and work of judges for the challenge
- Bi-monthly lunch events to showcase and critique projects
- Planning and execution of Challenge Award event October 2017
Research Conference (assumes venue is donated), 75 attendees, two days: Sub-total: $24,550
- Background research memos: 5 memos, 40 hours each @ $25 an hour = $5000
- Speaker Travel related expenses needing travel reimbursement for 10 people
- Airfare 6 @ $500 = $3000; 4 @ $1000= $4000
- Hotel @ $250 night x3 = $7500
- Ground transportation, parking and incidentals: $800
- Catered lunch: 75 attendees, 2 days, $20 each = $3000
- Afternoon coffee break: 75 attendees, 2 days, $5 each = $750
- Badges, printing and miscellaneous expenses: $500
Day of Code (assumes venue is donated), 50 attendees, one day: Sub-total: $1250
- Lunch for 50 attendees: $20 * 50 = $1000
- Coffee break for 50 attendees: $5 * 50 = $250
Development of the Research Commons Sub-total: $5200
- Web design: $1000
- Web development (e.g. customized Mediawiki deployment): $3000
- Software hosting and support (e.g. Discourse): $100/month = $1200
I School Challenge: Sub-total: $10,900
- Launch event/communications ?
- Lunch showcases: 6 events * $50 = $900
- Prize for best project: $10,000
Local: We will use mailing lists, email newsletters and announcements on the Web to contact potential participants at the School of Information, at UC Berkeley and across the San Francisco Bay Area. More targeted invitations will be used to reach representatives from relevant government agencies, academics in several relevant disciplines and employees and volunteers who work on or contribute to several online platforms, including Wikipedia and other Wikimedia projects. A diverse group of stakeholders available locally will be especially important for in-person events (research conference and Day of Code).
Global: In preparing for the research conference, we will seek to find participants to interview (remotely) from different Wiki projects, to better understand the needs of different user groups in, for example, different countries or different socioeconomic situations. If possible, we will offer video-conferencing for remote participation in events, including for key speakers from other countries.
The Research Commons will be designed as an online, collaborative resource in order to include contributions from outside the Bay Area and the US. We would welcome help in outreach to Wikimedia chapters and user groups, particularly outside the United States, to identify research and tools and help build a community for ongoing engagement. We will also work with contacts at UC Berkeley to reach out to fellow academics and researchers in other countries.
- A panel discussion in April 2016, "Toward a Gender-Inclusive Internet: Strategies to Counter Harassment, Revenge Porn, Threats, and Online Abuse" brought together experts working on or researching different platforms, including Wikimedia
- Reading group on online harassment, October 2015
- countering-harassment mailing list at UC Berkeley, set up for discussion of related projects/events
Code of conduct
CTSP activities are run using the CTSP Event Code of Conduct, which asks that participants are respectful and constructive and provides people to contact in the event of problems. (This is an example of a Friendly Space Policy.)
The Center for Technology, Society & Policy (CTSP) and the UC Berkeley School of Information (I School) can provide an institutional home both to maintain resources developed during the course of this project and to build on those results for future research and events.
Literature reviews, interview results and reports from the research conference and Day of Code can be maintained on UC Berkeley servers. CTSP and the I School can also provide basic ongoing maintenance of Mediawiki and discussion software for the research commons resource. To the extent that events and outreach can create a community of volunteers, online volunteers and Berkeley students will be encouraged to contribute to maintaining the repository of research over time.
If we are successful in demonstrating a year-long challenge project within and outside UC Berkeley, we may be able to identify in subsequent years issues affecting online communities that would also be amenable to focused, research-based work.
A research commons resource is useful for collecting research and identifying gaps in work, which can provide guidance to future students and to other researchers, so that projects can continue to develop in interventions to counter online harassment.
Measures of success
Long term measures of success:
- Over time, there is more relevant and actionable research on key questions related to online harassment that leads to new tools and solutions.
- A global community of researchers and technologists regularly contribute research and knowledge to the research commons and there is more exchange and collaboration across disciplines and across borders.
- Internet users are afforded more tools and strategies to address harassment.
- New technical solutions are developed and new actionable research is produced by the I School community and the broader Berkeley community as a result of the challenge.
- A research agenda for online harassment is developed which is widely supported and disseminated.
- The research commons is launched in 2017, providing the first global forum for knowledge exchange on the issue of online harassment.
Please use this section to tell us more about who is working on this project. For each member of the team, please describe any project-related skills, experience, or other background you have that might help contribute to making this idea a success.
- User:Npdoty: Director of the Center for Technology, Society & Policy at the UC Berkeley School of Information; Wikipedia editor since 2008; former software engineer at Microsoft; PhD student studying how privacy and other values are embedded in technical design.
- User:Leslieh96: former president/CEO of the Center for Democracy and Technology; Principal of the Harris Strategy Group; Lecturer and consultant, UC Berkeley School of Information; tech policy lawyer, Internet freedom and women's rights advocate.
- Volunteer Industry research + intervention ideas 220.127.116.11 16:55, 15 August 2016 (UTC)
Please paste links below to where relevant communities have been notified of your proposal, and to any other relevant community discussions. Need notification tips?
- post to Wikipedia Village pump (idea lab)
- wiki-research-l mailing list
- #wikimedia-research IRC channel (logs)
Do you think this project should be selected for an Individual Engagement Grant? Please add your name and rationale for endorsing this project below! (Other constructive feedback is welcome on the discussion page).
- I really think we should be looking into ways to improve online harassment. Online environments should be as safe if not safer than offline. Mitar (talk) 22:08, 11 August 2016 (UTC)
- This project aims to tackle an extremely important and relevant issue, that I am personally passionate about. As a woman in a field dominated by men, I have faced abuse in specialized online spaces like Stack Overflow, and on more general platforms like Reddit and Twitter. All along, I did not have the right tools and frameworks to address it/counter it.
- Although abuse is as much a problem offline, it has been further perpetuated by anonymity and the exalted position that "free speech" holds in most social media platforms like Reddit. The problem is, at what point does free speech become abuse? As an alum, I believe that the UC Berkeley I School has the right expertise to study this problem, and design interventions to tackle it. Ashwinisriram (talk) 20:26, 15 August 2016 (UTC)
- This is a thoughtful, considered approach to helping solve an important problem. I really like the focus on community engagement in combination with high quality research using an holistic approach - this is an absolute necessity for projects that aim towards solving normative problems. Kudos.--Hfordsa (talk) 09:44, 16 August 2016 (UTC)
- This proposal is timely and it is crafted in such a way so as to support a range of already existing - but, unfortunately, too often disconnected projects and research. The Berkeley I School and the CTSP are particularly well-suited to bringing folks together and building community in this space. One output I'd like to see, however, is a better understanding of the range of individuals (be they people, professionals, users, companies, etc...) for which online harassment should be a concern and way. Put another way, harassment on particular platforms should obviously be an issue developers/maintainers of that platform, but it could also be construed as a matter of research ethics for researchers who use data from that same platform. In this way, we can gain a better understanding of how a range of actors might be responsible for thinking about this problem (albeit in different ways). I think the collaborative nature of this proposal offers an opportunity to think in this expansive way. -Anna Lauren Hoffmann 18:45, 31 August 2016 (UTC)