The Foundations of People-Centric Digital Public Infrastructure
At the heart of JanDPI lies a set of principles we hold to be non-negotiable.
These are not aspirations to be pursued when convenient, nor checkboxes to be ticked at the end of a design process. They are the foundational commitments that distinguish a Digital Public Infrastructure that serves the public from one that merely operates in public.
Across two decades of grassroots work, from rural panchayats to international policy forums, the Digital Empowerment Foundation has learned that the difference between digital systems that empower and those that exclude often comes down to the principles on which they are built. Where principles are clear, design follows. Where principles are absent, exclusion fills the gap.
These are the principles that must guide the next generation of DPI in India and across the Global South. They speak to what DPI must be, and equally importantly, to how access must be ensured.
DPI MUST BE
Eight non-negotiable commitments for any infrastructure that calls itself “public.”
I. Government-Led & Publicly Accountable
Public infrastructure demands public stewardship.
Digital Public Infrastructure is, by its very name, a public undertaking. It cannot be quietly privatised through opaque partnerships, nor can its accountability be outsourced to terms of service that no citizen has read or agreed to in any meaningful sense.
We affirm that the state must lead. The government must own the rails, set the rules, and bear ultimate responsibility for ensuring that digital infrastructure serves the public interest. Private sector participation is welcome indeed, often necessary to bring innovation, efficiency, and capacity. But participation is not the same as ownership, and partnership is not the same as control.
When something goes wrong, and in any system at scale, something will go wrong, the citizen’s recourse must be to a public institution, not a help desk. The accountability must rest where the power rests: with the state, answerable to the people. Anything less is not public infrastructure; it is private infrastructure operating in public space.
II. A Public Good
Designed for the people, by the people, with the people.
DPI is not a product to be marketed, nor a platform to be monetised. It is a public good in the same family as roads, water systems, public education, and public health. Like these classical public goods, it must be available to all, accessible to all, and accountable to all.
This principle has profound design implications. A public good is not optimised for engagement or revenue; it is optimised for universal, equitable access. It is not designed for those who can pay the most or click the most; it is designed for those who need it the most. Its success is measured not in transactions processed, but in lives meaningfully served.
To say DPI is a public good is to commit, fundamentally, to a particular philosophy of design, one that treats every citizen as a rights-holder, not a user; as a participant, not a data point; as a co-creator, not a captured audience.
III. People-Centric
Grounded in lived experiences and bottom-up design.
The phrase “user-centric” has become so common in the technology industry that it has lost much of its meaning. JanDPI insists on something more rigorous and more demanding: people-centric design.
People-centric design begins not with assumptions about users, but with the lived realities of human beings, particularly those whose realities are most easily overlooked. It begins in the village, not in the boardroom. It begins with the woman who cannot read the screen, the elderly farmer whose fingerprints no longer authenticate, the migrant worker without a fixed address, the person with a disability for whom default interfaces are inaccessible by design.
Bottom-up design means that the people who will be most affected by a system have a meaningful voice in how it is built. Their experiences shape the requirements. Their feedback shapes the iteration. Their dignity shapes the defaults. This is not consultation as decoration; it is co-creation as discipline.
When systems are designed in this way, they tend to work better for everyone, including those for whom they were not specifically designed. Designing for the margins improves the centre. Designing only for the centre abandons the margins.
IV. Open & Interoperable
A common digital language without forcing uniformity.
Citizens should not have to navigate a maze of disconnected apps, logins, and platforms simply to access the public services they are entitled to. They should not have to remember a dozen passwords, switch between unrelated portals, or repeatedly submit the same information to different agencies. The complexity of bureaucracy is hard enough; digital systems must reduce it, not multiply it.
Open and interoperable DPI ensures that systems built by different agencies, governments, and providers can communicate seamlessly, sharing data, recognising authentications, and supporting transactions across boundaries. It allows new services to be built on existing rails without reinventing them. It enables a vibrant ecosystem of innovation without fragmenting the citizen’s experience.
But interoperability is not the same as uniformity. Just as people speak a common language while retaining their unique accents, dialects, and expressions, digital systems must communicate clearly while respecting local context, language, and need. Openness is the discipline that makes this possible through open standards, open APIs, open documentation, and open governance.
V. Secure & Privacy-Preserving
Data is not currency. Data is dignity.
In the digital age, personal data has too often been treated as a tradeable asset — collected by default, monetised by design, and protected only as an afterthought. We reject this framing entirely. Personal data is an extension of personhood, and its protection is a non-negotiable expression of respect for the citizen.
Every system in the DPI stack must be built with strong, demonstrable safeguards. Purpose limitation must be more than a clause in a privacy policy; it must be enforced in code. Consent must be informed, granular, and revocable, not buried in pages of legalese that no one reads. Data minimisation must be a design discipline: collect only what is necessary, retain it only as long as needed, and delete it when its purpose is served.
Security and privacy are particularly important precisely because the citizens DPI serves are often those least able to defend themselves against breaches, fraud, or surveillance. The poorest, the least literate, and the most marginalised are also the most vulnerable to digital harm. A truly people-centric DPI takes this asymmetry seriously and builds protection into the foundation, not the footnotes.
VI. Financially Inclusive
Real economic participation, not just access on paper.
The promise of digital financial infrastructure is profound: that every person, regardless of geography, income, or status, can be brought into the formal financial system as a full participant. India’s financial inclusion journey through Jan Dhan accounts, Aadhaar-enabled payments, and UPI represents one of the most ambitious such efforts in human history.
Yet inclusion measured only in accounts opened or apps downloaded tells an incomplete story. True financial inclusion is measured in lives transformed in the artisan who can finally access fair credit, the farmer who receives direct benefit transfers without leakage, the woman who controls her own savings for the first time, the small entrepreneur who can transact with anyone, anywhere.
People-centric DPI commits to financial inclusion that is real, not nominal. This means systems that are usable by those without smartphones or steady connectivity. It means transaction costs that do not penalise the poor. It means credit, insurance, and savings products that reach the informal economy with dignity. It means making sure that being “included” actually translates into being economically empowered, not merely registered.
VII. Accountable
Every system will fail someone. What matters is what happens next.
No system, however well-designed, will work for every person every time. Authentication will fail. Records will be incorrect. Benefits will be denied. Transactions will go wrong. The question is not whether such failures will occur, but whether the system has the integrity to address them when they do.
A truly people-centric DPI commits to clear, accessible, human grievance redressal mechanisms. This means a citizen who has been wrongly excluded must have a place to go — not a 1-800 number that never answers, not a portal that requires the very access that has just failed, but a real, local, responsive institution that can resolve the problem. It means timelines that are honoured, processes that are transparent, and outcomes that can be appealed.
Accountability also means that grievances are not treated as exceptions but as signals — data points that reveal where systems are falling short and must be improved. A DPI that listens to its grievances and responds to them is a DPI that gets better over time. A DPI that suppresses or ignores them is one that fails its public test, no matter how impressive its dashboards.
VIII. Meaningfully Accessible
Beyond technical availability usable in real lives.
There is a profound difference between availability and access. A service that exists somewhere in theory is not the same as a service that can be used by the person who needs it. A system that is technically accessible to anyone with a smartphone, broadband connection, digital literacy, valid documents, and the time to navigate complex interfaces is, in practice, accessible only to a privileged minority.
Meaningful access is access that works in the actual circumstances of the person who needs it in their language, at their literacy level, with their available device, within their physical reach, and at a cost they can afford. It is access that does not require the citizen to overcome the very disadvantages the system is meant to address.
A DPI that is “available” only on paper, or only to those who already have advantages, is not infrastructure at all. It is a privilege wearing the uniform of a public good. People-centric DPI rejects this performance of access and commits to the harder, more honest work of ensuring that the systems built to serve people actually do.
ACCESS MUST BE
Four commitments that translate principles into reality.
The principles above describe what DPI must be. But principles, however well-articulated, mean nothing if they do not translate into how citizens actually experience these systems on the ground. The following four commitments define what access must look like for a DPI to be genuinely people-centric.
Assisted
Supported by human intermediaries where needed.
The dominant assumption in digital design is that users will navigate systems on their own, equipped with the literacy, devices, time, and confidence to figure things out independently. For a significant portion of India’s population, this assumption is not merely flawed; it is actively excluded.
People-centric DPI recognises that self-service is a privilege, not a baseline. For those who are illiterate, semi-literate, elderly, disabled, first-time users, or simply unfamiliar with digital interfaces, the presence of a human intermediary is not an inefficiency to be eliminated but a service to be guaranteed.
Assisted access can take many forms: trained Soochnapreneurs in Community Information Resource Centres, Common Service Centres staffed by skilled operators, frontline workers like ASHAs equipped to help citizens navigate digital systems, panchayat-level digital assistants, or peer-support networks within communities. What unites these models is a recognition that the human element is not a fallback for digital failure — it is a feature of inclusive digital design.
Affordable & Reliable
No cost or connectivity barriers.
Public infrastructure must be affordable, or it is not public. A road that requires a toll only some can pay is no longer a public road. A digital service that requires data, devices, and downloads beyond the reach of those it claims to serve is not a public service in any meaningful sense.
Affordability cuts across many dimensions: the cost of devices, the cost of data, the cost of transactions, the cost of travel to access points, the cost of intermediaries, and the opportunity cost of time spent navigating broken systems. People-centric DPI takes all these costs seriously and works to minimise them — not by passing them on invisibly, but by addressing them through public investment, subsidy where appropriate, and design that respects the realities of low-income users.
Reliability is the twin of affordability. A system that works only when servers are responsive, networks are stable, and authentication infrastructure is functional cannot be the foundation on which entitlements depend. For the citizen who has walked five kilometres to a ration shop, “the server is down” is not a technical issue. It is a denial of food. Reliability, especially at the last mile, must be treated as a moral imperative, not a service-level agreement.
Multilingual & Inclusive by Design
Designed for every language and ability.
India is a country of extraordinary linguistic diversity — 22 scheduled languages, hundreds of mother tongues, and countless dialects. Yet the dominant language of digital interfaces remains English, with Hindi as a frequent secondary option and other Indian languages often treated as afterthoughts. This linguistic asymmetry replicates and deepens existing patterns of exclusion.
People-centric DPI insists that inclusion is not a feature added later; it is the foundation laid first. This means designing in multiple Indian languages from the start — not as translation projects bolted on after launch, but as core design considerations from the first wireframe. It means voice interfaces for those who cannot read, audio cues for those who cannot see, and tactile interfaces where appropriate. It means recognising that “accessibility” includes not only the categories formally defined in disability legislation, but also the wider realities of low literacy, unfamiliar interfaces, and cognitive load.
Inclusive design is, finally, good design for everyone. Systems built with the diversity of India in mind tend to be more robust, more flexible, and more humane — not just for those they were specifically designed to include, but for every citizen who encounters them.
Offline & Low-Tech Friendly
Available through low-tech and offline pathways.
Until every household in India has reliable broadband connectivity, no citizen’s rights can depend on it. As long as connectivity is uneven — and despite remarkable progress, it remains so — public infrastructure must work even where the network does not.
People-centric DPI is committed to offline-first and low-tech pathways as legitimate, dignified ways to access public services. This means feature-phone compatibility, SMS-based services, IVR systems, paper-digital hybrid processes, offline authentication where security permits, and local caching of critical data where appropriate. It means recognising that millions of Indians will continue to rely on basic technology for years to come, and designing systems that meet them there rather than waiting for them to upgrade.
Offline and low-tech access is not a step backward. It is a step toward universalism — a recognition that public infrastructure must serve all citizens in the conditions they actually live in, not the conditions designers wish they lived in.