Rapid Response Research (RRR) projects are quickly deployed scholarly interventions in pressing political, social, and cultural crises. Together, teams of researchers, technologists, librarians, faculty, and students can pool their existing skills and knowledges to make swift and thoughtful contributions through digital scholarship in these times of crisis. The temporality of a rapid response is relative and will vary depending on the situation, from a matter of days, to a week, or several weeks. Our model below is relevant to the variable timelines a situation might require, but it bears remembering that a crisis itself has an immediacy, and that RRR projects, accordingly, bring with them a pressure to respond with intensity and speed. Torn Apart/Separados is an example of RRR. While the recommendations below are based on RRR data narratives, many elements could be more broadly applicable to other types of RRR.
Another way of looking at RRR is as a new intervention in the fourth estate, between long-term scholarly commitments and the press. Closer to investigative reporting, but without shedding its scholarly frameworks, RRR breaks through several obstacles of contemporary journalism: a) RRR is not pressured to provide “two sides” to the story; b) despite its speed, RRR does not have as high a demand on turnaround as the press during a crisis; c) RRR can focus on a single project for a sustained period of time, as opposed to the competing demands on major journalistic data teams; d) we do not need to compete for scoops.
A critical first step is determining whether a particular crisis would benefit from RRR. It’s easy to jump to a techno-solutionist conclusion without considering its ramifications or fully understanding the problem. It’s essential to begin by exploring the topic quickly and in great detail to identify possibilities for intervention and to consider the possible consequences of an intervention. We must be self-reflexive about our best intentions to ensure we aren’t hurting, rather than helping.
Some questions and actions to consider are:
If, at the end of that very early stage of research, you realize that you should not go forward with a project, stop. No shame in stopping, even if you have already started public conversations about a possible project. Share the reasons why you stopped with your community. That, in the end, may be your best contribution to a crisis.
If you have explored the situation and concluded that RRR would be beneficial, the next step is assembling a team. As with any collaborative project, the composition of the team is crucial to success. Given the pressure on a team undertaking RRR, it’s essential that team members know what’s ahead of them.
When assembling a team or deciding to join one, consider:
Recognizing that there may be those who are interested in participating but won’t work at the speed necessary for RRR, it can be useful to differentiate between a core team, contributors, and consultants, and to manage expectations accordingly:
Core team: Core team members are primarily responsible for the intellectual leadership, design, and implementation of the project. They are prepared to work apace each other (though not necessarily synchronously), remain in regular contact about their availability, and to communicate clearly with the group about their progress and need for support or assistance.
Contributors: Contributors are an essential part of undertaking RRR. They are willing to assist and offer valuable insight on the project even though they may not be able to work at the rapid pace of the core team. Contributors are welcomed and encouraged to participate in all processes and conversations along with the core team—research, design, and implementation—and to share their expertise. They are typically on “stand-by,” willing to pick up small-scale but critical jobs delegated by the core team. These could include taking on small data wrangling and coding tasks, monitoring news and social media cycles, doing research, creating bibliographies, managing crowdsourcing, providing password-protected access to data, translating, and running DIY security detail to limit the team’s exposure to harassment online once the project goes live.
Consultants: Another indispensable part of an RRR team is the cadre of colleagues you consult with about the project. These could range from librarians to social workers to faculty to technologists, and more. Exactly who you need to consult will arise organically from the needs of the project. For RRR, it’s critical to remember that “expertise” is not the sole domain of the academy, and the people working on the ground in response to a crisis may understand the situation much better than those who aren’t.
Organization and leadership on the project will arise organically from the core team in response to the needs and shape of the project. Self-reliance is key, though. Generally, each member of the core team needs to take responsibility and leadership for at least one part of the project—though often more— across its lifespan. This could range from data management to project scope to technical components of the project to project management to translation, and so on. In addition to their domain leadership, they pitch in as time and skills allow on other dimensions of the project. Having a relatively flat, non-hierarchical team structure allows each core team member to leverage their strengths collaboratively to move the project forward.
In order to ensure proper credit, and to keep trains running on time, the people playing the role of project manager should keep a spreadsheet updated with assigned roles as they emerge and shift throughout the life of the project. One tab can be dedicated to active roles, and another to stand-bys. Each row in the spreadsheet represents a person. Columns can include contact info, initial point of contact, roles, known strengths, pronouns, and other relevant information.
With a team in place, it’s essential to dedicate time to an initial intensive research phase. This research will build on the foundation established during the process of deciding whether to undertake RRR. The purpose of this research phase is twofold: to continue understanding the crisis and its implications and to identify data sources that will be used in the project. While all of your team members might not be undertaking this research, it’s important that they are kept apprised of the findings and understand the background of the crisis and relevant data.
Drawing on existing data sets is integral to working at speed. Useful sources for data sets include publicly available government data, requests through information freedom laws (such as the Freedom of Information Act in the United States), non-governmental organizations, and open data repositories. Recognizing, however, that even the most “official” of sources are far from objective, it’s essential to evaluate the data with a critical eye, considering:
At least one member of your team—ideally more—should be responsible for thoroughly wrapping their mind around the data sets that will be used in the project. This becomes crucial as you move into the implementation phase of the project, particularly as other members of the team start working with the data sets. The people who clearly understand the data will be able to offer advice and guidance on how the data can and cannot be used or named, and they will have an intuitive sense of when visualizations seem inconsistent with the data.
With data sets in hand, and a clear understanding of their affordances and limitations, you can move into implementation. Working under the time constraints of a crisis project, it’s necessary for the team to remain agile in thinking, pacing, planning, and design, while maintaining clear, consistent channels of conversation.
There is no funding. Unlike projects that rely on grant funding, RRR relies on the good will and intensity of purpose of committed volunteers stealing time away from their personal lives and day jobs—or folding it into their day jobs, whenever possible—to come together for short-term collaboration in rapid response to a crisis. Broad support from institutional administrators can go a long way to clear up time, and the Nimble Tents Toolkit exists in part to garner that support. That said, our model assumes that you may not have such support. In this temporary coalition, team members commit to the project and nothing more.
The voluntary nature of the project speaks to the need for careful credit and attribution of team members, both in the project’s credits and in any use of the project in presentations or publications in the future. For the latter, all team members should be acknowledged, and the team member should distinguish between their own work (“what I did”) and the work of other team members (“what they did”).
To facilitate communication, the team should agree on a messaging platform they will regularly check. We recommend Telegram or Signal—because of their encryption—but other options may be useful. One recommended feature is the ability to easily create sub-groups for focused conversations. Another useful feature is the ability to drop files and links in the chat box for easy retrieval.
A nimble but clear division of labor is necessary to undertaking RRR. As tasks are completed, roles can change. While a task is being completed, though, that role should be clearly defined.
Implementation of one visualization (or other type of narrative) should begin as soon as it’s conceived, even while the rest of the team continues with the research and data wrangling. As good ideas continue to surface, other team members can transition to implementing them. Prototype as you go to test ideas before committing to them, but recognize that part of a nimble workflow is being open to final products that look different than you may have initially intended. Eventually, the general research process ends, and everyone shifts to working on implementation tasks that may include the sort of research associated with smaller tasks, such as filling out or expanding new data sets.
A good idea should be recognized as such by most members of the team depending on the goals. These may be recognized as newsworthy, insightful, simple, surprising, or pressing. When considering ideas emerging within the team, keep in mind the multiple audiences that may encounter your project and the responses you hope to elicit. Such audiences may include the general public; activists, lawyers, politicians, or others in the position to intervene in the crisis; and academics, among others. Successful RRR will appeal to these wildly different audiences by design. Some considerations include:
This last point speaks to the ethical challenges of undertaking RRR. There may be instances where you have some data you shouldn’t share or make easily available because a well-intentioned but uninformed audience might rush to action without fully understanding the situation or thinking through the consequences. Revisit the question of what information should and shouldn’t be shared frequently throughout your work.
To negotiate these challenges, include a narrative in the project that describes your team’s process, from research to implementation (a process more likely to be iterative than linear). In addition to framing the project as a scholarly intervention, this narrative will help your audience understand the project and your team’s motivations. Describe the challenges of working with your data sets and their limits, decisions made in response to the team’s ethical commitments, and the types of visualizations your audiences will encounter. The narrative is also an opportunity to add some dimension or texture to the experience of responding to the crisis, whether by highlighting strange material encountered during the research project that may not have found its way into other parts of the project or addressing the rabbit holes and dead ends in the process.
Considering the main publishing mechanism for the results of your RRR will likely be the open web—though less public distribution mechanisms can be imagined, depending on the situation—the minutiae of web development will be very relevant to your team’s speed. Here are a few considerations that may help your team move quickly through development:
Special attention should be paid to an effective Data Management strategy. Moacir P. de Sá Pereira has prepared a separate section, “Nimble Data Management,” that lays out a model with implications beyond RRR.
Some common components that facilitate the development process:
During the implementation phase, it’s also important to continue paying attention to mainstream and social media responses to the crisis. In such moments, the news cycle moves quickly, the social media outrage cycle follows suit, and, inevitably, sometimes both are wrong. At least one or two members of the team should be adept at culling information through media firehoses—Twitter lists, search engine alerts, RSS feeds, simultaneous browser tabs, etc.—or should coordinate with a designated media monitor. The rush to report and to respond to a crisis can lead to misreporting and belated corrections. If your team has been doing their research, it’s entirely possible you will be ahead of the news cycle.
The media can also be useful contacts, particularly if you are both willing to share data. In light of their need to protect their sources or their pressure to “break” a story, they may not be willing or may regard your efforts with suspicion, but it’s worth leveraging your team’s contacts and social media connections to ask. It’s even more important to build relationships and make contacts with people working on the ground in response to a crisis because they understand the situation better than anyone else. However, keep in mind that they have their hands full with their work and may not have time or energy to respond quickly.
In the early stages of the project, consider sharing with your networks on social media that you are working on RRR. This is not only a mechanism for building a committed and skilled team but also helps plant early seeds of interest within your professional communities. If you started with an interdisciplinary and interprofessional team, make sure early team members do as much across their overlapping communities. As the project progresses, make sure to check in publicly once in a while on social media with a teaser or preview of things to come. These teasers are great places to park parts of the project that may not end up in the final version, but which are relevant to the overall conversation.
One of the goals of these early public moments is to start courting the attention of the media. Depending on the news cycle and your pre-launch activities on social media, the press may want to know about your project early on and help you get access to mainstream publics. Consider also reaching out to your local newspapers or your institution’s communications office. To facilitate this process, start drafting a press release early, so core members of the team with media experience can start adding highlights and talking points to share with the press. This helps you frame the story and ensure that proper credit is given to all members of the core team. The press traditionally needs a hero or two to make stories “human.” This works against the collaborative ethos for RRR and de-incentivizes participation and goodwill on a team. A press release can help mitigate some of that by providing a frame and rhetorical tricks-of-the-trade that can help reporters deal with non-hierarchical team formations.
Give yourself several hours for everyone on the core team to see the completed project pre-publication. Some minor tweaks will inevitably arise from this last minute review, so it’s important to have most, if not all, hands on deck in the final stretch. If you have colleagues outside of your team poised to amplify your launch on social media, make sure to communicate with them about delays so they don’t inadvertently scoop you while your team is working on finishing touches.
Here’s a checklist of items you should have in place to make sure you are ready for what’s next:
If all systems are ready, activate the production URL a couple hours prior to announcement for final checks. Designate one person to be the announcer and link sharer, and quickly have everyone else follow suit in sharing and re-sharing. Godspeed!
If the project is well-received by your immediate public, or if your work was covered by a media outlet, you will probably get (more) media requests. The project’s main liaisons can handle incoming requests and ensure that all team members who so desire receive equal media exposure. We recommend that you find a balance between seasoned media speakers in your team and exposure for members with less media experience. Most reporters will want to tell a positive story about your project and be as inspired by your work as the general public. In other words, they probably want to see it succeed too. This means that they will be amenable to requests that fit your particular professional incentives (e.g. naming your employer or giving you a chance to speak about your other work).
Besides the press, you will inevitably engage in a dialogue with your colleagues. Perhaps you should consider harnessing some of the feedback from them through a formal single-blind or open peer-review mechanism.
Chances are you will be asked for your curated datasets. Make sure to have a clear data sharing policy in place. Once you do, make sure you keep your open data in an accessible and stable repository for sharing. Private data should be readily available in common formats to share over secure channels.
While RRR is a short-term affiliation intended to respond to a crisis, it’s likely that the conclusion of the project will bring with it additional questions that went unanswered, ideas that could not feasibly be implemented within time constraints, or interest from others wishing to be involved with the project. This is an opportunity to pass on your model of RRR and your datasets to a new team. We suggest sharing this guide with them and taking time to explain the particularities of your project, with attention to the ethical dimension of the data, the values of the team, and potential avenues of research that your team identified but that were outside the scope of your project.
As our experiences show, wading into the waters of public information in moments of crises is a decision that shouldn’t be undertaken lightly. First and foremost, it must be done with the safety of affected communities in mind. Researchers must also be aware that this work can be taxing on the spirit. Exposure to high levels of cruelty and injustice, for example, are probably best handled by those who feel prepared to fix their eyes on intimate horrors and still fight on with clarity of mind and a stout will. We urge you take care of yourself, regardless of the situation, at the same time as you care for others—through the work. (Note adapted from “Textures” in Torn Apart/Separados).
