Ushahidi and our community is known for broaching new ground. To accompany this, we are also keen to encourage more research. With that in mind, we're excited to share two important reports from the Knight Foundation and Crowdglobe Funded by Internews.
For the past few months, I have been summarizing Ushahidi community successes and best practices into sound-byte form for those TLDR (Too Long Didn't Read) people:
5 T's to a successful Ushahidi deployment
Now for the readers
This is going to be a long post, because the lessons learned highlighted in each report are the same that deployers experience. Plus, they are key to our planning as Ushahidi continues to meet the community needs and build good software. Happy reading and distilling!
Both reports highlight the need for planning and preparedness for your Ushahidi projects. The research methodologies and software were different. The Knight report included interviews with deployers and focused on the Ushahidi platform. The Crowdglobe report was a both quantitative and qualitative research report focused on the Crowdmap software. Our team has been working to address some of the metrics and feedback collected from both the reports.
I think we need to honor those who tinker before launching full projects. Helping deployers move from tinkering to full projects takes a good portion of my attention. It is a layered learning process that includes huge topics like digital literacy and security, access to technology, technical acumen, outreach planning, Ushahidi software knowledge, and basic project management. There are staff and community mentors who try to assist whenever possible. Our wiki is full of tips and tools. We welcome your input to share best practices.
To academics and researchers: we would like to match you with deployers to do more research. See our wiki page on how to get started. We often post about funding opportunities for research on the community mailing list too.
Experiments in Media Innovation
The Nieman Lab did a review of the report:
The News Challenge asks that projects not just use open source tools, but also release the software they create to the public. Of the 2009 winners, DocumentCloud, Ushahidi, and Data Visualization stood out for how widely their code has been embraced. In particular, the report says those projects found favor because they persistently released updates rather than waiting for a finished product. They also released code that could be used outside of their specific app.
Key lessons learned cited in the Knight report
Ushahidi has been successful in creating a platform widely used to organize, visualize and map crowdsourced information. The team's efforts to scale their platform, release a cloud-based service, and partner with nonprofits to deploy the tools on the ground in Kenya contain a number of useful lessons.
- Community Capacity: Ushahidi's platform requires technical acumen to install, launch and maintain. Local organizations using the platform, however, often did not make the necessary provisions to outsource technical assistance to integrate the software into their projects. Organizations that conducted a tech assessment before starting a project were much better placed to understand and determine their internal technology needs, the capacity of their partners, and their ability to successfully utilize the software. To help with the challenges of implementing the software, Ushahidi has provided training and demonstrations to prospective testers, organized Ushahidi 101 events, facilitated a space for organizations to match up with local IT talent, and developed its community resources page and user listserv to connect current and potential users.
- Designing with Bandwidth Constraints: In several deployments in Kenya, individuals did not often have sufficient bandwidth and consistent Internet connectivity to easily upload reports on the Ushahidi platform. With this in mind, many users have encouraged the creation of a "light" version of the Ushahidi website to enable low-bandwidth users to access the information and mapping systems. More recently, an option was created that allows users to collect data offline and then upload the information later when an Internet connection becomes available.
- Safety and Privacy in the Crowd: For future projects in high-risk security environments, more time should be spent planning to ensure the safety and privacy of individuals using the Ushahidi platform. This might include training volunteers on the importance of privacy, crafting public messages to the media, using back-up volunteers from the diaspora or others who can process information remotely, and establishing backup servers in external location so that operations can continue at all times.
- Print Still Matters: In some of the Ushahidi deployments, individuals had difficulty using the Ushahidi platform because of a lack of high-speed bandwidth and their unfamiliarity navigating and interacting with web-based maps (e.g. how to scroll through a map and zoom in and out). In instances where technology and digital literacy skills are limited, printed paper versions of Ushahidi-generated web-based maps play a critical role in helping communities use crowdsourced information.
CrowdGlobe is an initiative highlighting deployments and research in the Information and Communication Technology (ICT) field on how to strategically and effectively use mobile and mapping technology for new media, emergency response, political engagement and civic programs. We're honored that Crowdglobe and Internews did their first report on data collected from our Crowdmap service -- our hosted Ushahidi service. The report examined the use of nearly 13,000 instances. The methodology examined general user patterns through quantitative impact alongside user surveys and highlighted case studies.
highlights from the Crowdglobe report
Data for this report was collected in October 2011:
- The Numbers: Our work took advantage of statistical analysis, quantitative content analysis and exploratory surveys. The quantitative analysis revealed that 93% of the 12,000+ Crowdmaps analyzed had fewer than 10 reports while 61% of Crowdmaps were identical to the default Crowdmap setting, i.e., they had not been customized or used at all."
- The Challenges: Some of the biggest challenges cited by users of the Ushahidi platform included "keeping reports up to date, embedding pictures and documents;" "the need to display the data on something other than a map;" "getting the word out quickly after the launch of the site;" and "getting people to submit reports." When asked about failures, respondents' answers ranged quite widely, from persistent technical problems to mobilizing volunteer involvement. Some of the biggest challenges cited by users of the Ushahidi platform included "keeping reports up to date, embedding pictures and documents;" "the need to display the data on something other than a map;" "getting the word out quickly after the launch of the site;" and "getting people to submit reports." When asked about failures, respondents' answers ranged quite widely, from persistent technical problems to mobilizing volunteer involvement. Many noted that using the platform was simply too time-consuming.
- Survey Research: A separate survey for semi-structured interviews was developed for users who launched high-profile projects using the self-hosted Ushahidi platform. A total of 37 high-profile projects were identified for the survey and seven respondents completed the survey, i.e., ~19% response rate. Each of these users represented a formal organization. All seven users represented formal organizations who responded to the second survey had used the platform in response to a complex humanitarian emergency or "natural" disaster, with two deployments being in developing countries and the remainder in the "Global South." Some of the biggest challenges cited by users of the Ushahidi platform included "keeping reports up to date, embedding pictures and documents;" "the need to display the data on something other than a map;" "getting the word out quickly after the launch of the site;" and "getting people to submit reports." When asked about failures, respondents' answers ranged quite widely, from persistent technical problems to mobilizing volunteer involvement. Many noted that using the platform was simply too time-consuming.
- Next Steps: As we assess the growth and impact of Ushahidi in general and crowdsourcing in particular we should keep in mind that we are still at the very start of a transformative process.
- The Big Takeaway: This takes us to what is, in a sense, our most important preliminary conclusion. This project's greatest contribution might come in the form of establishing best practices for users and potential users, and perhaps even new procedures at Ushahidi. For example, despite the surprisingly high levels of education many responded had, many still reported that the complexity of the platform impeded their full use of it. Some did not even understand the necessity of public awareness and the need for public mobilization. This suggests that one important corrective measure is found in further refinement and ease of use for even novices.
Thank you to Knight Foundation and Internews for continuing to support the journey of our community and Ushahidi.
Also, if you are still looking for more to read, I've been digging into Joseph Bock's The Technology of Nonviolence: Social Media and violence prevention."
Heather Leson has a diverse technical background and experience in digital response and open source communities. She brings a passion for community building and idea hacking to her role as Director Community Engagement at Ushahidi. With a formal education in politics and library information technology, she has over 15 years of experience in technical incident management, software life cycle development, customer care and communications in Internet technologies. A leader within open communities, including Random Hacks of Kindness and CrisisCommons, Heather is the organizer of numerous successful unconference and hackathon events. Bridging information and people inspires her.
A version of this post originally appeared on the Ushahidi blog.