DFS’ Executive Director Ewan Oglethorpe on his journey from Silicon Valley to Kathmandu and the formation of Data Friendly Space.
What was your role in the data sector before you embraced the humanitarian sector?
Before making my way to the humanitarian sector, I worked in the US as an academic researcher and in the private tech sector. As a researcher, I studied how student retention could be optimized based on results of various surveys given to students. Later, I worked as a Data Scientist and Data Engineer for Teradata Aster and Stitch Fix which provided me experience in both established corporations and startup environments.
You landed in Kathmandu to lend a helping hand during the 2015 Nepal Earthquake. You started working with the UN information management team; what was your first impression of the situation, and how different was it from your silicon valley experiences?
I’d been traveling to Nepal for 3 years prior to the earthquake and fell in love with the country and its people. After that disaster, everything was radically different from what I’d experienced before. I saw a lot of people with good intentions attempting to solve a pressing humanitarian crisis. But as a data scientist the way data was being collected, stored and analyzed concerned me. A lot of the technical IT facets of the response lagged far behind what I had learned from the private tech sector. I felt like I could help.
How did the formation of DFS take shape? What were your initial goals, and where do you see the organization now?
I had initially planned to go to Nepal for just a month but I was so struck by the acute need for improved data systems in the humanitarian community that one month turned into about three years. DFS took shape from the experience in Nepal. My initial goal was simply to create tools and services that would help to modernize the humanitarian data ecosystem, and bring it closer to what I had been seeing in the private sector. Today, the organization has grown into this vision more than I ever thought possible.
We are now a team of more than 85 people spread across six continents composed of humanitarian assessment experts, data analysts, IT whizzes and innovation leaders. The teams in Nepal that have partnered with us since the early days have continued to grow and this has provided quality employment to the younger Nepali generation, which is an aspect of the work that I’m most proud of. We have executed projects in partnership and on behalf of humanitarian organizations and donors including UNHCR, IFRC, DFID, USAID BHA, IDMC, UN Environment and the Danish Refugee Council to name a few. While I’d like to see DFS continue on this path, going forward I think it’s important for us to focus more on projects that have direct impact among the humanitarian beneficiary populations we serve.
What are the recent humanitarian projects where your organization (DFS) has significantly contributed?
I’ve enjoyed seeing DFS be involved both in projects that have taken a few weeks to complete and also ones that we’ve been working on for years, and envision continuing to do so in the future. Our biggest project has certainly been DEEP, the first humanitarian project I undertook and one the organization has grown around. Through DEEP, DFS is currently providing secondary data review and analysis to 40 countries around the world. For instance, we are currently supporting the Humanitarian Needs Overview process in Afghanistan, Somalia, Sudan and South Sudan and providing support to the refugee crisis in Central and South America. Outside of the traditional humanitarian scope in which we work, DFS has supported Transparency International for the past two years in transitioning to a new database solution used by their national societies to manage reported incidents of corruption.
It is essential to have an open-source data system in the humanitarian sector, but how does an organization like DFS help humanitarian organizations ensure that the data are correctly used and their security is not compromised?
There’s a strong argument that security can actually be improved through open source technology since having a software’s code visible to the rest of the world forces developers to deliver to a higher standard, and for the community to find bugs and provide fixes. DFS regularly puts our applications through third party security audits to ensure we’re following industry leading, best practices when it comes to cybersecurity. In terms of data use, whenever we share data publicly, we follow policies that shield potentially sensitive or personally identifiable information and make the humanitarian purpose of the datasets explicitly stated.
How are the latest developments in technology such as Machine Learning aiding the humanitarian efforts?
I think that the potential for improvement of both speed and accuracy of humanitarian action through ML/NLP techniques is quite substantial, and largely underutilized at the moment. There are a lot of simple technologies that can be implemented that will provide immediate gains. Furthermore, DFS is developing more complex solutions that will advance our platforms for years to come. The challenge isn’t necessarily the creation of these solutions, but more so the adoption and use of emerging technologies by the humanitarian community. I ultimately see these tools fueling a more robust strategy from data sourcing to collation/annotation and finally to analysis/visualization. Software development feels piecemeal, but I believe that several years from now we’ll look back on the manual ways humanitarian responders did our work and wonder how we ever managed.
How do you see the role of the private sector coming together with humanitarian organizations to fight poverty, climate crisis, war, pandemics?
I’ll admit that it sounds idealistic, but at the end of the day we’re all in this together. I think that the private sector and humanitarian sector should be very closely aligned in this mission. The private sector has an inordinate amount of knowledge, expertise and key technologies that can be shared with the humanitarian sector to ameliorate responses. I’ve been pleased with a number of initiatives I’ve seen from private sector tech companies, but I believe there is still a long way to go. While DFS is a nonprofit, and I firmly believe in the power of nonprofits, private sector organizations can also deliver social good if it’s something they put their efforts towards.
How do you manage to run an organization with teams spread across six continents?
By never really knowing what time it is (it doesn’t help that our partner organizations in Nepal boast a time zone that is 15 minutes off the international standard), and obscure sleeping patterns. In all seriousness, I would say that the core of how we operate is mutual trust and responsibility across all levels of our team. Due to time differences and physical distances, micromanagement simply isn’t possible and communication needs to be very intentional and clear. We hire people and partner with organizations that are mature and can properly deliver to support our humanitarian mission without much hand holding.
Ewan Oglethorpe is the Executive Director at Data Friendly Space.
To learn more about DFS, its projects and DEEP, please sign up for our newsletter on www.datafriendlyspace.com
Download DFS organization profile here.
Or write to us at firstname.lastname@example.org