Ethics in the Use of ICT in Development Projects: An Interview with Jennifer Chan

Jennifer Chan, MD/MPH, is an Associate Faculty Member of the Harvard Humanitarian Initiative (HHI) and Lead Instructor for Humanitarian Technologies at the Humanitarian Academy at Harvard University. She is also an Assistant Professor and Director of Global Emergency Medicine at Northwestern University’s Feinberg School of Medicine. Her recent activities have focused on humanitarian technologies and crisis mapping with a focus on field operations. As a consultant, she helps evaluate open source technology organizations such as Ushahidi, trains emerging practitioners in humanitarian technologies, and researches the interface between humanitarian agencies and volunteer & technical communities.

Q: Can you tell us a bit about your work at the intersection of technology and humanitarian assistance?

A: The area of work I’ve focused on over the past couple of years is commonly referred to as crisis mapping in the field of humanitarian technology. Humanitarian technology refers to the application of technology to address humanitarian issues, including both acute onset disasters and complex ongoing humanitarian emergencies.

I’ve worked in high-tech, low-tech and near no-tech settings, and this has pushed me to continually think about and re-explore the relationship between information, privacy, security, and ethics. As a researcher and consultant I’ve worked with NGOs and other agencies to help them strategically and programmatically integrate ICT into complex humanitarian situations, including disasters, conflict settings, and drought prone regions. This work has included work on when [the use of technology] is appropriate, and how it is safe to use.

Q: From an ethics perspective, what do groups designing development projects with a technology component need to keep in mind?

A: A key ethical perspective to keep in mind is the  “do no harm” principle. Codes of conduct, when revised and applied to crisis mapping projects, provide very important guidance but this is still a work in progress. A concise, practical, and widely accepted set of ethical guidelines, and best practices that take into account principles, standards or guidelines, have yet to be developed.

ICT projects have potential for faster, more accurate and wider inclusion of information, but they also present vulnerabilities that can cause risk and potential harm. Information shared using technology can be accidentally misrepresented or interpreted, carelessly distributed without consent, and even intentionally manipulated for a multitude of purposes that are not in line with the good intentions of the ICT project. And there are additional vulnerabilities that lie within the technology itself. But the flip side is that these types of obstacles are frequently surmountable with a good framework for learning, iteration, and revisions embedded in the project.

Working with technology, it is helpful to recognize and accept the tenet that access to information is related in many ways to power. Understanding that first is key. Beyond thinking about how technology can influence the design of a project, organizations should seek to understand the existing information ecosystem by asking:  What decisions would be made, or how would behavior change, on the basis of the information gathered with technology? Where are the vulnerabilities of information collection, sharing and communication that exist with the intended users and beneficiaries of the project, and with the technology itself?

In my evaluations of these types of projects, frequently I see organizations thinking about how they’ll use technology before developing a nuanced understanding of what information is desired, or how the information collected would be used. Rather infrequently is there a comprehensive assessment of information as it relates to privacy, risk and harm.  I often recommend that this process be inverted and efforts made to include as many users in the discussion. This speaks to the importance of user-centered design.

Q: What best practice have you seen in the use of ICT in development projects that incorporates sound ethical principles or guidelines?

A: In my experience, many of the emerging ethical principles and guidelines in these projects are focused on information, security and privacy. On the design side, organizations need to be mindful of the various planning phases and how much time it takes to understand the information security and privacy context. There needs to be an assessment at the country level of what information and privacy guidelines need to be considered. In addition, organizations should outline steps for the project, ask stakeholders to recognize and abide by codes of conduct, and link these activities to ethical principles that relate to ICT as well as to the organization’s mission.

It’s also critical to engage stakeholders at the local level to understand what is important to consider when it comes to information and privacy. In certain contexts a change in region within a country could mean a significant change in risk. Engaging the right people in the design process is essential to developing a sound security protocol. This can require a lot of resources, planning and advocacy for adequate funds to achieve these important goals, but will serve the project better over the long term.

Q: What about the importance of enabling a user to “opt in” or “opt out” in the context of a humanitarian mapping project?

A: With any crisis mapping project, the first thing you need to do is consider the purpose integrating ICT. But before you collect information you need to ask yourself and the community members involved: Why use a map? What’s the anticipated purpose? And within this exercise also ask whether or not this has a potential to do harm in addition to good.

Ideally, you would reach out to all of the people in the information ecosystem. You would ask local communities: This is what we’d like to do, what do you think about it? Find out what are people’s feelings are. Is this ability to share information beneficial or uncomfortable to them? These conversations become telling about whether there are under recognized or unanticipated challenges.

Q: Do ethical considerations change depending on whether you’re dealing with an area where there is high access to technology versus one with low or no access?

A: To some degree, yes, and to some degree no. What it really comes down to is an understanding of the information flow. You could have online-based mapping or curation of crowdsourced information, but if that information goes into a print format and is shared with communities, or if meetings take information from an online source and then folks talk about it, then you’re looking at an online-offline information pattern. Sharing information in an offline environment does not necessarily remove the risks associated with it.

To ensure information security and privacy, it’s useful for organizations to map out entire information communications systems, from high- to low- and no-tech communications, and then build in a vulnerability and risk model that engages the community in this process to make this a participatory practice. Once you have this framework, then challenge the assumptions around the risk model.  Could there be risk in collecting information by itself, even in an offline environment?  Having this framework facilitates community engagement, and allows analysis at both global to local levels.

Q: What else is important to consider?

A: Scenario planning is important from a programmatic design standpoint.  Whether you’re using FrontlineSMS or another ICT platform, scenario planning and simulation are valuable to understand vulnerabilities, risk, and unexpected privacy issues. It also can demonstrate how much time is required to start – often much more than people realize.

Having a budget line item for a simulation and practicum period is also very useful. Taking policy level discussions into simulation to implement protocols and understand what happens in the last mile helps strengthen the implementation phase of the program. You can learn from mistakes and ideally mitigate any potential negative effects with beneficiaries. Finally, simulations and a practicum period enable people involved in the project time to learn the tool and, more importantly, the process, and hopefully come to a deeper understanding the how ICT fits within the larger project goal.  Organizations that do this are much more successful than many of the implementations we’re seeing.

Q: What is the significance of organizations like United Methodist Communications incorporating ICT4D into their project design?

A: It’s important that groups thinking about engaging in ICT related activities, particularly those that are new to it, to remember that no matter what tool or technology approach you take ultimately these interventions come down to access to information and the benefit and the power associated with it. That’s why it’s important to have a thoughtful process around what that means, and what access to information entails for different people. We frequently assume that there are inherent benefits associated with increasing access to ICT – but are there also risks associated with it and harm as well?

We have to remember that when it comes to privacy and security in crisis mapping or ICT4D, this is a work in progress. Every organization should be part of the larger conversation about what works and what doesn’t—not only as listener, but as contributor. There’s still a ways to go.

—

For more information about best practice in privacy and security in ICT implementations, see:

Chamales, George, Lea Shanley, and Aaron Lovell. Towards Trustworthy Social Media and Crowdsourcing. The Wilson Center, Washington, D.C. 2013

Collaborative Learning Projects. The Do No Harm Handbook (the framework for analyzing the impact of assistance in conflict). Collaborative for Development Action, Inc., Cambridge, MA. 2004.

The International Committee of the Red Cross. Professional Standards for Protection Work. 2013 Edition. Chapter 6, Pg 77.

Building Human-Centered Design into ICT4D Projects: An Interview with Danny Alexander and Sean Hewens of IDEO.org

Sean Hewens is knowledge manager and in-house counsel at IDEO.org. He oversees IDEO.org’s mission to spread human-centered design throughout the social sector. In his role as Knowledge Manager, Sean has helped document and tell stories about every aspect of the IDEO.org design process. Previously, Sean worked at a law firm and then ran a non-profit building computer labs in Kenya and Tanzania.

Danny Alexander is a Senior Designer at IDEO.org. Prior to IDEO.org he worked in industrial and product design, and  has been a partner in several start-ups. He works with IDEO.org’s clients and partners to explore new frontiers in design, with a focus on the intersection of design and social good.

Q: What is human-centered design?

A: [Sean]: Human-­‐centered design is a problem-solving process that puts humans at the very center. There are three key tenets.

First, immerse yourself within the community as much as possible [to understand the problem]. You can start very specific with folks in the community. Learn from them as much as possible.

Second, you focus on brainstorming and prototyping rapidly. The idea with brainstorming is to go as broad as you can with ideas. Use what you learned from the community and ideate.

Third, you get those ideas back out to the community to get feedback from real users as quickly as possible in the form of prototypes. Our early prototypes often fail, but in a way that allows us to iterate and refine our ideas based upon feedback from real users.

Q: Why is human-centered design important in the social sector, and is the social sector lagging behind the private sector in this regard?

A: [Sean]: Human-centered design started with IDEO in the 1980s designing the computer mouse and the folding design for the laptop computer. As IDEO expanded its process, it began working with the social sector and found human-centered design to be very effective.

There are a couple of reasons for this. In the social sector you have funders and beneficiaries. Frequently you find donor-driven programs and solutions that aren’t based on the real world needs of actual users, or the communities that are supposed to be benefitted. Human-centered design is an effective tool that designs programs and solutions based on real user needs.

[Danny]: There are a lot of missing feedback loops in social sector implementations. The private sector has a feedback loop, although it may be one-dimensional: do people buy a product, yes or no? But in international development you have projects being implemented thousands of miles away from where decisions are made. Frequently, there’s no feedback loop so it’s hard to say: Is it working, and are people choosing to use this? In these instances you have programs that are not at all sustainable and not having real impact, and yet donors continue to think they’re successful and promote them just because they have been implemented. Human-centered design helps give voice to the community, and ensures design is built around user needs and is sustainable.

[Sean]: Before I joined IDEO.org, when I was building a computer lab in Tanzania for my non-profit, from a tech perspective we nailed it. But we didn’t consider the actual needs of the community until we arrived there to build the computer lab. We didn’t ask: what about the parents of the kids who will use the computer lab? What about teachers or community members who won’t have access to the lab? All of these reasons that our project was most likely to fail were rooted around people and the community. Unfortunately, most of what we actually considered before we went was the technology of the computer lab that our donors had paid for us to build. This was my first glimpse into the idea that starting with the users in the community is the much more sustainable approach.

Q: Why is human-centered design important to the field of ICT4D?

A: [Sean]: In general the development community is very risk averse. If things go badly, you’re not talking about losing quarterly profits but losing lives. One of the benefits of human-centered design is to mitigate risk by testing early and failing fast.

In the developing world so often where projects fail is the not around the actual technology or solution, but about how it is implemented in the context of the community. You need to ask yourself: Even if it the technology is good, how it might fail anyway? For example, is microcredit needed to enable the community to make necessary purchase? Is training needed to ensure that the technology will be used? You need to figure out all the conditions necessary to the technology being implemented and sustainable.

[Danny]: In our work, we encourage testing ideas in a rough, unfinished form, often before the technology is even developed.

One example is a recent project called Pump Away, funded by the Bill and Melinda Gates Foundation as part of their Reinvent the Toilet initiative, which is investing in technologies to address sanitation challenges around the world. Our work on Pump Away was to take one of the technologies they’ve developed—an Omni‐Ingestor—and develop the business, service and brand for a sustainable pit latrine emptying service in Zambia. The team spent a few weeks on the ground in Zambia, interviewing community members about what the sanitation challenges were. This made things really tangible and enabled the team to prototype quickly, even though the technology is still months, or years, from being finished.

One of the key criteria for the project was to design a service that would be financially sustainable, and that investors would invest in. So we asked ourselves: What kind of a business model will make people pay to use this, and what makes it better than the competition? Could we get a bunch of people to sign up to empty latrines at the same time in the same place for greater efficiency?

To test this idea, we dressed our local translator up as a salesman and gave him an artificial price for this service and then let him loose in the community to see if people would sign up. Within two hours, we had more demand than the technology could even fulfill. Rapid prototyping enabled us to do all of this before anything was built. Testing in this fast, rough way enabled us to prove the concept and understand local demand. We also learned that people were interested in emptying part of their latrine if they couldn’t afford to empty the full latrine each time.

In the context of ICT4D, human-centered design can help with the design of a technology, and the context around it, long before the technology is ready for launch.

Q: Is failure at certain times not only acceptable but important?

A: [Sean]: Our ethos is not that failure is good or bad, but that when you learn from it, failure can be a very positive part of the process. You want to try to get some of the failing out early so that you can learn from it and let it influence the design of a better more successful project.

[Danny]: One example of failure with a small “f” is the Clean Team project in Ghana. In early field research, the team showed community members different kinds of toilets that they might get through the program to learn what their aspirational ideal was from a theoretical perspective. Most people said the standard western flush toilet was the ideal.

So a few weeks later the team went back to Ghana with 5 or 6 full-scale prototypes, some of which were water flush toilets that would work without connection to a water system or to a sewage grid. Despite having designed the toilets based on the community’s aspirational ideal, the water flush toilet was an abject failure. Two days later every single one had overflowed.

Why? In this particular community people used water to cleanse instead of paper. Older people couldn’t figure out how to use it. And then people realized they had to pay for the water. Across the board, the feedback was negative.

So what did we learn? When we interviewed the community, that’s what they said they wanted, but just asking wasn’t good enough. We needed to build, test, and fail. Trust me, you learn more from cleaning up a dirty toilet than you ever will from qualitative work or a survey. Prototyping and testing enabled us to fail on the scale of just a few households, whereas if we had invested in a major program around this aspirational toilet there would have been thousands of homes with overflowing toilets.

[Sean]: Here’s an opposite example of Failure with a big “F”, where folks didn’t fail early. In this example, an organization spent a lot of time investing in a large program that was rolled out without ever testing with the community. The program, which was implemented in a refugee camp in northern Kenya [near the border with Somalia], provided a nutritional pack called Sprinkles. The idea was that users would open the vitamin packet and sprinkle it on food or in water to provide basic vitamins for kids. But when the program was implemented, none of the parents would give it to their kids. Why? The foil wrappers that contained the vitamins looked just like condom wrappers. So this program had spent all this money designing and distributing the packets, and then failed. If they had run a small pilot initially, or even prototyped the packaging, they would have failed in a small way earlier in the process, and been able to change the packaging at a much smaller cost.

You want to start small, pilot early on and then scale up. This is the value of human- centered design.

–­

For more information on applying human-centered design to the work of social enterprises and NGOs, download the free IDEO.org toolkit available here.

Using SMS to Strengthen Community Communications in the Democratic Republic of the Congo

Betty Musau of The United Methodist Church talks about how cellphone technology is strengthening communications and transforming communities in the Democratic Republic of the Congo, where she is a UMC conference communicator.

“Instead of waiting for Sunday or Wednesday for an announcement in Church, now we use SMS to tell people” about urgent news using FrontlineSMS, Betty says.

What happens as a result of increased community communications via SMS? “People are changing their behavior,” such as by boiling water during a cholera outbreak, or using bednets to prevent malaria.

“Cellphones are the only means of communication that are effective because people don’t have access to the internet,” Betty says.

What’s her vision for the future? “If every woman can have access to a cellphone, it will save lives and transform the community.”

Learn more about how The United Methodist Church is using technology to improve lives, livelihoods and spiritual wellbeing at http://www.umcom.org, and follow the discussion on Twitter using #ICT4DBP.