FOSS United letter to NPCI requesting open sourcing of UPI

[Venkatesh_Hariharan]

We sent the letter below to the CEO of the National Payment Corporation of India (NPCI)

Venky

I am writing to you on behalf of FOSS United, a non-profit organization that promotes Free and Open Source Software (FOSS) in India and aims at promoting and strengthening the FOSS ecosystem in India. We have been tracking the growth of UPI in India with great interest. We firmly believe that UPI can be a flagship technology project to emerge out of India to global acclaim and use. In this note, we request NPCI to open source UPI and outline a few steps for doing so. We will be happy to have a more detailed discussion with you and your team, and offer our knowledge and expertise for this effort.

While India is known for its prowess in software services, there have been very few Indian technologies that have created a global impact. In the aftermath of Covid, many governments have realized the urgency of having robust, universal, and widely accessible digital payments that cover a whole spectrum of payment use cases including peer-to-peer, merchant, and Government to People (G2P) payments. Since UPI is not only a massive success story already but a case study that is now eliciting global interest, we request that NPCI open source UPI (its specifications and software artifacts) and make it a Digital Public Good (DPG). This will pave the way for UPI to play a key role in setting up modern, digital payment systems globally, and to showcase India’s growing technological prowess on the global stage. By open sourcing UPI, it can positively impact the lives of billions of people across the globe, garner tremendous goodwill for India, and potentially build a global instant payment network around an Indian standard.

This document suggests a few first steps that could be taken to ensure that UPI can be open sourced with appropriate checks and balances so that due credit accrues to India and NPCI.

Legal: The UPI specifications in the beginning can be released under a Creative Commons Attribution Non Derivative (CC-BY-ND) license that gives the UPI Open Source Software (UPIOSS) project the sole rights to evolve the UPI specifications. This can be enforced through branding and trademarks. For example, the term “open source” is a trademarked term and only those licenses that have been officially approved by the Open Source Initiative (OSI) which owns this trademark can be called an open source license. Trademarking combined with the CC-BY-ND license will give UPIOSS control over the UPI specifications.

Organizational: The success of UPI has attracted great global interest from countries and companies around the world who see the advantages of having public rails for payments infrastructure. For example, Google recently wrote to the US Federal Reserve Board highlighting the example of UPI-based digital payment in India in order to build “FedNow” – a real-time gross settlement service (RTGS) for faster digital payments in the US.

The UPIOSS project can enroll leading national, corporate, and community adopters into a consortium that will implement, sustain and evolve the UPI project to greater heights. Given the growing interesting in a modern payments infrastructure like UPI, this consortium can raise capital from multilateral development agencies and its membership base to build a world class open source community and ecosystem, evolve the UPI specifications, and rapidly accelerate the deployment of UPI around the world, reinforcing India’s role in building UPI.

Leadership: Given the global interest in UPI, it should be straightforward to build a strong global open source community around UPI. Countries such as Singapore, UAE, and France that are already working with UPI could become part of the UPIOSS project. The knowledge capital, and more importantly, the direct, hands-on experience with massive scale that NPCI has in creating and deploying UPI, is very hard to replicate. This affords NPCI the de facto position of being able to provide strong support and leadership to UPIOSS, making it THE global community for a modern payments infrastructure.

Risks in not open sourcing UPI

There are big risks to not open sourcing UPI. Open Source is now the de-facto global software development and innovation model. Many countries choose open source software as the basis for their foundational infrastructure to protect their digital sovereignty and avoid vendor lock-in. If other countries or corporations develop open source instant payment systems that become popular, the window of opportunity for global UPI adoption, despite NPCI pioneering it, will be lost.

The open source model of collaborative innovation leads to rapid technological development. The best example of this is Linux which now runs on everything from the tiniest embedded computers to smartwatches to smartphones to desktops and servers and supercomputers and has even gone to Mars aboard the Mars Rover. Such astonishing breadth of innovation would not have been possible if Linux was not open source. Given the rapid pace of change, any technology that does not evolve and gain widespread acceptance is like an ice cube in the desert; its value could shrink rapidly. By open sourcing UPI, we can enable communities to rapidly adapt it to local contexts by adding multilingual user interfaces, customizing to low-resource contexts like feature phones and many others.

Given the global interest in G2P and UPI, we suggest that NPCI should quickly put together an open source consortium around UPI. We are confident that UPI can go from being an Indian success story into a global standard. By open sourcing UPI, we can create a globally adopted flagship project out of India. We request time with you and your team to discuss this further.

With warm regards,

Venkatesh Hariharan
Public Policy Director
FOSS United

[rushabh]

@Venkatesh_Hariharan - thanks for sharing this. Putting out some of my reservations here for the record as well.

While it makes sense for NPCI to open source UPI components / protocols, I think we should also think about the architecture this promotes.

This whole wave of “Digital Public Goods” and “India Stack” will end up hepling centralisation of state power more than anything else. The assumption here is that the state is and will continue to be benevolent.

While we are against big headwinds, we should also think about alternate architectures that the state should use that will promote maximum autonomy and freedom rather than tight control. The state should define models of co-operation rather than run services that run deep into society.

[Venkatesh_Hariharan]

Rushabh, I agree with you on your “centralization of state power” argument. Concentration of power (whether in the public sector or the private sector) leads to bad outcomes. There is also no guarantee that the state’s intent is benevolent/will always remain benevolent. We have seen democracies come close to autocracies, and this is a dance that is playing out in many parts of the world. Also agree on individual “autonomy and freedom.” Would love to understand what you mean when you say, “models of cooperation.” At the IndiaFOSS 3.0 panel discussion one of the points made was that the state should create an enabling framework for citizens to build solutions and Taiwan was cited as an example. Is that what you mean?

[rushabh]

@Venkatesh_Hariharan

We have seen democracies come close to autocracies, and this is a dance that is playing out in many parts of the world. Also agree on individual “autonomy and freedom.” Would love to understand what you mean when you say, “models of cooperation.”

I think we need to revolt against this whole idea of “Digital Public Goods” which is essentially technology enablement for the government. At best this will make government “delivery” more efficient, at worst, it will be used to curb political dissent (like many governments across the spectrum in the past several years have demonstrated).

We need to ask ourselves deeply - “What do we expect our government to do for us”. Blindly volunteering to make the state more powerful by “Digital Public Goods” seems very short sighted.

I think community services should be "local "and be owned by civil society rather than massive services by government. Even ONDC is a non-starter unless the state arm-twists or mandates large companies to comply. What we need from the government is just to “stay away”. Digital commerce, education, health etc should be left to the civil society to innovate. If the state dominates these sectors, they will essentially end up deepening the control of a few industrialists on society. Large government contracts in garb of DPI get awarded to companies like TCS, Infosys, Google etc. Follow the money. If the state really wants to control big-tech, it needs to strengthen monopoly laws and break up some of these companies.

We should be very wary of associating FOSS United with such initiatives. At best we can be independent observers or ideally even push for less government intervention in areas like trade, commerce, health, education etc.

[Sai_Phanindra]

Some really great points about DPI, UPI, DPG and more in this panel discussion: https://www.youtube.com/watch?v=G1_Qk1hcKfI Someone in the audience also pointed out (at 45:00) that it is actually “Digital public services infrastructure” and they shortened it to DPI for marketing purposes.