Genesis and Vision of OPAL
OPAL came out of the recognition that leveraging ‘big data’ sources collected by private companies for research and policy purposes has been a conundrum, for legitimate ethical and commercial reasons.
To date, one of the most valuable of these data sources, known as Call Detail Records (CDRs) collected by telecom operators for billing purposes, have been accessed and analyzed externally either through data challenges such as the Orange’s D4D Challenges or through bilateral contracts and arrangements. These types of engagements have offered ample evidence of the promise and demand.
A large body of academic literature has showed how computational analysis of CDRs (but also banking data), typically alongside traditional survey data and other official statistics, can help capture socio-economic outcomes and processes at high levels of geographic and temporal granularities and degrees of complexities—including disease spread, poverty, literacy, crime, as well as optimize public service delivery and transportation systems, notably.
At the same time, there are risks associated with using such personal, connected data. Concerns for privacy and security have grown as the notion of data anonymization was increasingly tested, as evidenced by research that has come out of MIT, and the nature and extent of the surveillance activities of the US National Security Agency, fueling fears of an Orwellian future. The prospects of growing imbalances between groups that have access to the data and capacities and those who do not, and the resulting concentration of power, are also worrisome. A related criticism is that algorithms increasingly used to make policy decisions are akin to ‘black boxes’ concealing rules and procedures that cannot be subjected to public scrutiny and redress.
In light of these obstacles and requirements, OPAL’s vision is the development of a new type of techno-institutional system built on trust that leverages private sector data to foster transparency, agility, accountability and inclusion while respecting privacy and security. To that end, OPAL will reflect and foster a paradigmatic shift to turn Big Data on its head and "save it from itself". The first step is to send the algorithms to the data, not the other way around, so that data are never exposed to theft and misuse. The second step is to co-design how big data algorithms are developed and used, so that they served local needs and respect local standards, instead of imposing external perspectives and expertise.
Dual Components of OPAL
OPAL combines two complementary components to create a holistic, self-sustaining system:
The technology track includes an open platform and open algorithms with beta testers.
The governance track manages the participatory design process, an orientation and ethics committee (CODE), and capacity building programs.
These dual components are being implemented in two initial pilot countries, Senegal and Colombia, with funding from the French Development Agency (Agence française de développement, AFD). In partnership with their National Statistical Offices—respectively ANSD and DANE—, as well as Colombia's National Planning Department (DNP), and leading local telecom operators—Orange-Sonatel and Telefónica Colombia. OPAL has established partnerships with other organizations, including Friendly User Testers (FUTs) in both countries.
At the time of its formal launch in Q4 2018, OPAL includes:
In order to unlock the potential of private data for public good, OPAL provides access to a unique suite of assets, services, and expertise that be leveraged collectively to drive shared outcomes. OPAL enables current actors in the space to better leverage their data or systems for greater results.
By “sending the code to the data” rather than the other way around, OPAL seeks to address the current challenges in this system and spur dialogues and develop data services on the basis of greater trust between all parties involved.
This system operates as follows:
Partner private companies (a telecom operator, for example) allow OPAL to access its servers through a secured platform.
The governance system including a Council for the Orientations of Development and Ethics (CODE) ensures that the algorithms and use cases are ethically sound, context relevant, etc.; users benefit from capacity building activities.
Key indicators derived from private sector data such as population density, poverty levels, or mobility patterns, feed into use cases in various public policy and economic domains.
Certified open algorithms created by developers are sent to run on the servers of partner private companies, behind their firewalls.
Developed through participatory design, an orientation committee, and capacity building activities
An open platform and algorithms