DPG Charter: Key takeaways on deploying digital public goods as infrastructure

Throughout Summer 2022, the Digital Impact Alliance (DIAL) held a series of consultations in partnership with the Digital Public Goods Alliance (DPGA) and other partners to inform the scope and substance of the Digital Public Goods (DPG) Charter. Read the introductory, first, and second blogs of this series for background and context.


Summary

  • Insight #1: In practice, the adoption of digital public goods as foundational infrastructure begins with reforming procurement and tender policies.
    • Solution #1: Introduce digital public goods and infrastructure as part of digital transformation strategy positions and blueprints.
    • Solution #2: Develop and institutionalize open-source policies  to help grow a culture around the procurement and use of open-source technologies in government.
    • Solution #3: Reform the tendering system to better acknowledge the needs and capacity of systems integrators as partners.
    • Solution #4: Identify appropriate licensing requirements for open-source software used in procurement tenders.
    • Solution #5: Consider data models in project designs, so that systems integrators can improve how they bid on projects.
  • Insight #2: While they can either replace or complement existing government systems, the open and interoperable nature of digital public goods makes it easier to build platform-centric digital public infrastructure that works for government and citizens.
    • Solution #1: Allow the main contributors to open-source software to play an assurance and enablement role, in order to help get participation from systems integrators.
    • Solution #2: Ensure that the implementation, operations and maintenance of new systems are allocated appropriate budgets, and that the necessary bodies and institutional structures are set up and resourced.
    • Solution #3: Make a case to service providers and vendors for supporting all aspects of open-source deployment and maintenance, as well as integrations with systems and services.
    • Solution #4: Strengthen relationships with the private sector around open-source.
    • Solution #5: Use a whole-of-government approach to deliver on all of the use cases demanded by government, the private sector, and individuals.

 

Background

Due to their open-source nature, digital public goods can, in theory, help countries build digital public infrastructure more quickly, economically, and effectively than otherwise relying on proprietary solutions. That said, even as policymakers look to digital public goods as alternatives to proprietary solutions, adopting and implementing them can be difficult.

Implementation challenges — from procurement to design to maintenance — abound, especially since most countries have legacy systems already in place. To support implementation of safe, secure, and inclusive digital public infrastructure, the DPG Charter will mobilize commitments to ensure that countries have sufficient funding, technical capacity, strategies, and processes in place to build and scale end-to-end digital public infrastructure that addresses pressing national needs and empowers people, organizations, businesses, and civil society.

Our most recent consultations with product owners, global policymakers, systems integrators, and African policymakers repeatedly touched on challenges in these areas. In reflecting on these challenges, two key insights surfaced on Implementation, centered on the need to ensure countries can reform procurement and prioritize openness and interoperability.

———

Insight #1: In practice, the adoption and implementation of digital public goods as foundational infrastructure begins with reforming procurement and tender policies.

Every country has unique requirements and needs for their digital public infrastructure. Because of these requirements, as well as unique experiences and resource constraints, national governments end up experimenting with alternative approaches to building digital infrastructure and delivering public services digitally.

“So, [when] you buy an ERP–an enterprise resource planning system–from Oracle, they customize it for you, for your needs. But they also leave you with a lot more modules than what you asked for. However, you are not supposed to ever touch those modules; if you do, even accidentally, they will always make sure they catch you and fine you a huge amount of money. Which is normally a multiple of the normalized fee you paid.”

Vendor lock-in can constrain digital transformation in governments by imposing burdensome restrictions, limiting the flexibility to adapt and innovate, and giving countries less ownership over their digital and data capabilities.

In our consultations, numerous policymakers expressed frustration and exhaustion in their relationships with the large (usually multinational) software companies that were frequently responsible for their government’s digital infrastructure. Many agencies felt taken advantage of, often becoming responsible for solutions they were not interested in and subject to fees they felt helpless in the face of. This echoes a 2018 ID4Africa survey in which professionals at African identity authorities cited this as their top challenge.

Since many governments do not have the resources or power to litigate against large, multinational software companies, this made them beholden to restrictive agreements, pushing them to consider other potential pathways and solutions. Additionally, many noted how the issue of vendor lock-in has real implications for digital sovereignty — the idea that countries ought to have ownership and control over the hardware and software they use, as well as the data generated from the use of that technology.

There are also practical considerations. Years of suffering from onerous service contracts — including high fees for changes or upgrades, restrictions preventing switching or cancellation of services, and other hidden fees and charges — have led many governments to search for alternatives to proprietary software.

“Of course with proprietary [technology] there’s always a risk of vendor interest. It’s really important to understand the vendor’s business model because you really need to understand what you’re trying to do. You need to understand if it’s aligned with what you want to do, because if it’s not aligned when proprietary comes in, it just can’t support this [type of] model.”

Many countries report that they are now exploring how to use more open approaches to software, standards, and protocols that can eliminate some of the headaches they experience, and to enable more rapid experimentation and adaptation. That said, the current procurement policies in place in many governments have been, in part, responsible for the challenges governments are facing in the selection of digital public goods.

Echoing themes from DIAL’s research, commonly cited challenges for countries looking to reform their procurement and tender policies include (amongst other things): the technical framing of tenders and RFPs; varied terminology; lengthy processing and contracting arrangements; and challenges with discoverability and marketing of services. 

Our consultations uncovered several positive recommendations for improving the procurement of digital public goods at the country-level:

  • Introduce digital public goods and infrastructure as part of digital transformation strategy positions and blueprints. By including these elements as part of broader planning around digital transformation, governments have another policy lever to help standardize new business models for procuring and maintaining open technologies.
  • Develop and institutionalize open-source policies to help grow  a culture around the procurement and use of open-source technologies in government. These policies can help shape the enabling environment for open source, but must also be given more weighting in policy discussions, so that policymakers can make a robust case for — and provide evidence of — procuring open-source software over proprietary alternatives. Increasingly, governments and academic institutions are establishing Open Source Program Offices to cultivate and institutionalize a policy and procurement environment that enables governments to confidently adopt open solutions where appropriate. 
  • Reform the tendering system to better acknowledge the needs and capacity of systems integrators as partners. Systems integrators expressed a desire to better understand how governments seek to use open source and for what, so that they can better support them in adopting both types of solutions in their appropriate contexts. This will help governments address how the capabilities are evaluated when they are bidding.
  • Identify appropriate licensing requirements for open-source software used in procurement tenders. This will help systems integrators to understand the technology they are considering to deploy and put in appropriate bids.
  • Consider data models in project designs, so that systems integrators can improve how they bid on projects. Systems integrators must be able to understand how governments intend to govern the data generated from the digital public infrastructure they build. Considering these issues from the outset is critical for projects to be successful, and has implications for countries looking to exercise more digital and data sovereignty.

———

Insight #2: While they can either replace or complement existing government systems, the open and interoperable nature of digital public goods makes it easier to build platform-centric digital public infrastructure that works for government and citizens.

While there are numerous open alternatives to existing proprietary solutions (and others still being developed or even conceived), many of those in our consultations spoke about how countries pursue diverse paths for building their digital public infrastructure.

Open-source software and standards might one day wholly replace proprietary solutions in some (or even many) governments, but the option to work with the private sector will always exist. It is for this reason that as countries redesign or build new digital public infrastructure — using a combination of proprietary and open-source software, standards, protocols, and data hosting solutions — they must prioritize openness and interoperability by design, no matter the licensing.

“[O]pen-source and open standards are really critical to help reduce the risk of vendor lock-in, and that component-ized architecture can allow you to begin to plug and play different elements.”

As countries build this infrastructure, existing digital public goods are well-positioned to complement, rather than replace, existing government systems. Digital public goods are open by design, and many of them are designed (or can be designed) to be more open and interoperable than their proprietary counterparts.

As one participant in our consultations noted, building digital public infrastructure is often thought of in binary terms as “build or buy,” with both build and buy choices often leveraging proprietary options. A more effective approach examines the openness and interoperability of existing infrastructure and capabilities, and building more open digital ecosystems there. While not the norm, countries are experimenting with approaches that include a mix of ‘build and buy’ that works for their requirements and needs.

For this reason, those building that infrastructure need to have a clear understanding of their government’s needs, requirements, and capabilities in order to improve how they communicate with service providers, a need expressed frequently in our consultations. That said, the process of integrating legacy systems with new, more open systems is rarely straight forward.

Our consultations uncovered three key challenges which constrain countries in this area:

  • Blending open and proprietary solutions: Systems integrators — the companies that support implementation and maintenance — often do not have experience with blending different types of systems (open/proprietary, legacy/new). Building more open digital public infrastructure does not necessarily prescribe using open or proprietary solutions, but governments need more knowledge on how to blend the two in practice.
  • Lack of awareness: Many governments lack awareness of how to make their existing digital infrastructure work effectively with new technologies and products, many of which may be open source. Governments are almost never going to move wholesale to open source, and a focus on understanding and marketing to their existing needs and capabilities will help them to start building more open digital ecosystems.
  • Resource requirements: Implementation of new government systems — regardless of the underlying challenges — suffers from a failure to plan and resource for the long-term, which limits the ability of these systems to deliver results for companies and individuals. The business models  for developing and running open-source technology products need to be reformed to help these systems be sustainable for the long-term.

Our consultations also uncovered a few ideas for how to make this happen in practice:

  • Allow the main contributors to open-source software to play an assurance and enablement role, in order to help get participation from systems integrators. Examples of potential support functions these actors could provide include high-quality documentation, guides and training, videos, and learning materials.
  • Ensure that the implementation, operations and maintenance of new systems are allocated appropriate budgets, and that the necessary bodies and institutional structures are set up and resourced. While system integrators are still leveraged to drive implementations, these support bodies and structures can advise governments on the procurement, scoping, implementation, program design and operations, and management considerations, whether as part of the government or as an independent agency.
  • Make a case to service providers and vendors for supporting all aspects of open-source deployment and maintenance, as well as integrations with systems and services. At the moment, systems integrators that already understand the value of open-source and how to build relationships with government around it are still in the minority. Return-on-investment (ROI) must be part of the conversation around the sustainability and impact of these implementations.
  • Strengthen relationships with the private sector around open-source. There tends to be dissonance between governments and the private sector actors helping them to implement and deploy open-source technology, which can create downstream barriers (usually related to cost of implementation) that are passed on to citizens.
  • Use a whole-of-government approach to deliver on all of the use cases demanded by government, the private sector, and individuals. This requires overcoming existing silos, navigating entrenched organizational cultures, and elevating design and budget decisions to the executive branch. Creating this level of political will can be challenging and slow, depending on the context.

———

These insights and practical next steps are invaluable to catalyzing effective commitments through the DPG Charter, which imagines  a world where everyone has access to education,  health care, and the essential services they need not only to survive, but to thrive. Digital products, particularly those that are open-source and designed for the SDGs, can help us make that a reality, if we work together.

With immense gratitude to those who participated in the consultative process, we look forward to sharing more insights through this blog series with you in the weeks to come. Stay tuned!