Bridging the FLOSS divide between the North and the South

2026 Special edition - Free/libre and open source software: Visions of openness from the Global South
Attachment Size
Bridging the FLOSS divide 220.33 KB
Authored by
Penguin Selectah

Organization

Cloud68.co

Website

https://cloud68.co/

A divide that doesn't show up on the map

By most quantitative measures, free/libre and open source software (FLOSS) has never been less of a Northern affair. GitHub's Octoverse report describes a rapidly growing number of developers worldwide, with the fastest expansion in Africa, Latin America and Asia; India is projected to host the largest developer community on the platform within a few years, and countries like Brazil, Indonesia, Nigeria and Kenya post growth rates that Northern communities have not seen in a decade.1 The Eclipse Foundation's State of Open Source in the Global South, the first survey of its kind, garnered responses from 423 developers and policy makers across 18 countries. It found that 77% of respondents used open source software, 37% contributed to projects, 28% maintained them and 22% had created their own.2 People in the Global South are not just discovering FLOSS, as some may assume. They have been a part of the movement for decades, and now, in growing numbers, they are giving it a new lease of life.

And yet anyone who has spent time in FLOSS communities outside the wealthy core of North America and Western Europe knows that presence is not the same as power. The foundations that steward the most critical projects, the funders who decide which work is "infrastructure" and which is a hobby, the conferences where roadmaps are set, the maintainers whose approval merges a contribution: all remain overwhelmingly concentrated in the Global North. The result is a movement that is demographically Southern and institutionally Northern, a gap that runs through funding, governance, language, recognition and the basic material conditions of participation.

This report asks what constitutes that divide, and what bridging it would genuinely require. Two arguments run through it. First, the divide is not primarily one of geography or talent, but of position within power structures, which are typically centred in places the map calls "North". Second, bridging it cannot mean helping the South catch up to Northern norms. It means redistributing the power to define what FLOSS is for.

Beyond the binary: The view from the Balkans

Before dissecting the divide, it is worth complicating the categories. The North and South frame is indispensable shorthand, but it flattens as much as it reveals, and nowhere is that clearer than in the region where I was born: the Balkans (specifically the Western Balkans) or as European Union (EU) institutions have branded it over the last decade to avoid the "negative" connotation of the word Balkans, "Southeastern Europe".

Geographically, the Balkans is Europe. Structurally, its FLOSS communities live many of the conditions this edition associates with the Global South: a civil society dependent on donors, chronic brain drain, weak institutional uptake of open technologies, and a peripheral position relative to the funding and governance centres of the continent they belong to. The countries of the region sit at or outside the edges of EU programmes; their developers watch calls for proposals designed in Brussels, Berlin or Amsterdam, with eligibility criteria, fiscal requirements and application cultures built for organisations unlike theirs.

The communities themselves are anything but passive. In Kosovo, Free Libre Open Source Software Kosova (FLOSSK) has been active since 2009, running what grew into one of the region's most important free software conferences, contributing to the OpenStreetMap crisis mapping tool, translating software, and refurbishing laptops for schoolchildren.3 In Albania, Open Labs Hackerspace in Tirana had for over a decade anchored a volunteer community around events like the OSCAL conference, localisation work, and advocacy for openness in public institutions.4 Recent monitoring by the EU's Open Source Observatory notes that in Kosovo, the FLOSS ecosystem remains active despite the absence of any state actor systematically prioritising it: civil society carries what public institutions do not.5

But the fragility is equally well documented. Regional research by the Balkan Investigative Reporting Network (BIRN) on open data and digitalisation found that donors play a decisive role in the region's openness initiatives, and that platforms frequently stagnate once donor funding ends, because institutions lack the will or resources to sustain them independently.6 Communities run by volunteers burn out the same way: a hackerspace's continuity can hinge on a few people not emigrating in the same year, which was the case in Albania.

The Balkans, in other words, is a "South within the North", and its existence is analytically useful. It demonstrates that the FLOSS divide tracks economic and institutional peripherality, not latitude. Marginalised communities in wealthy countries face versions of the same barriers; privileged enclaves in poorer countries sometimes escape them. If we understand the divide this way, "bridging" stops meaning geographic charity and starts meaning structural redistribution wherever the periphery happens to be.

Anatomy of the divide

What, concretely, separates FLOSS practitioners in the periphery from those in the core? The barriers cluster into three layers: material, structural and cultural. They compound one another, which is why fixes aimed at a single issue so rarely work.

Material conditions

Open source's founding myth is the volunteer who contributes in their free time. That myth quietly assumes that free time exists, that electricity and connectivity are stable, that a laptop capable of compiling a modern codebase is affordable, and that unpaid work today can be traded for career capital tomorrow. None of these assumptions holds evenly across the world. Where bandwidth is metered and expensive, cloning a large repository is a financial decision. Where salaries are low and side work is a necessity, the subsidy of leisure that Northern contributors donate to projects simply is not available. Respondents to the Eclipse Foundation survey identified career opportunity as a central motivation for FLOSS participation,7 which is rational precisely because contribution is costly, and the cost must return something.

The material layer also includes the mundane brutality of mobility: visa regimes. FLOSS is coordinated online but governed in person, at summits, sprints and foundation meetings held overwhelmingly in North America and Western Europe. Contributors from most of Africa, Asia, Latin America and, until visa liberalisation, parts of the Balkans routinely face refusals, humiliating and expensive application processes, or costs exceeding a month's income to attend the events where decisions about "their" projects are made.

Structural position

The second layer concerns where the institutions sit. The foundations that act as fiscal hosts and governance bodies for critical projects are legally domiciled in the Global North – executive directors and other decision-making staff included. Funding follows the same geography: philanthropic and public money for digital infrastructure originates mostly in the United States and Western Europe, and even deliberately open programmes inherit the grammar of their origin: application languages, reporting standards, eligibility rules, assumptions about what an applicant organisation looks like.

Nadia Eghbal's influential research on the maintenance crisis regarding publicly available source code, Roads and Bridges, shows that the digital infrastructure everyone depends on is maintained by a small group of people with far too few resources.8 What that research understandably did not foreground is that the subsequent wave of funding it inspired flowed through Northern institutions to, largely, Northern maintainers, because they were legible to funders in ways peripheral contributors were not. Being legible, having a registered entity, a bank account acceptable to compliance departments, fluency in writing grants, references inside the network: this is itself capital, and it is unevenly distributed.

Maintainership follows structure too. Contribution statistics from the South are rising fast, but commit counts are not governance. The path from first pull request to maintainer to project leadership runs through trust networks, and trust networks are built at the conferences people cannot get visas for, in the time zones that put Asia-Pacific contributors in meetings at 3 a.m., and through the informal mentorship that flows most easily between people who resemble each other and share similar upbringings.

Culture and language

The third layer is the least visible and arguably the deepest. English is FLOSS's operating language: code comments, documentation, mailing lists, conference talks and funding applications. This is a gate that admits some and quietly filters others out, and it shapes whose problems get treated as general. A bug that affects users in an environment with poor bandwidth is "an edge case"; a feature needed in a European enterprise deployment is "core". Documentation gaps hit hardest where formal computer science education is scarcest and teaching yourself depends on written material.

Community norms are cultural artefacts as well. Northern FLOSS culture prizes blunt public critique, while promoting yourself and assertively claiming credit are styles that carry different social costs in different cultures, and different risks depending on one's gender, race and precarity. What reads as meritocracy from the centre often reads, from the periphery, as a set of unwritten house rules that one is expected to already know.

One finding of the Eclipse survey should complicate any deficit narrative, however. The survey found that women make up roughly 34% of FLOSS contributors in the Global South, against a global figure closer to 14%.9 On one of the movement's most notorious inclusion failures, the periphery is outperforming the centre. The South is not waiting to be taught diversity; in some respects it is already the teacher.

Who pays, who decides: Funding that imposes the way things are done

If one lever moves the whole divide, it is money. Not only how much, but how it travels.

Consider the fund this very publication is connected to. NGI Zero, financed by the European Commission's Next Generation Internet initiative, and led by NLnet, is by design one of the few accessible funding mechanisms FLOSS has had: small grants, a short application, no requirement to be a registered organisation, and a deliberate effort to support work that commercial funders ignore.10 It is, in the best sense, an experiment in low-threshold public funding. And yet its reach illustrates the problem: it is a European instrument, funded by European taxpayers with a mandate oriented to the European (or to be more correct, the EU) internet. Its openness travels only as far as awareness of it does, and awareness travels through networks. A developer in Novi Sad, Nairobi or Recife who has never heard of NLnet, has no peer who has received a grant, and has no experience decoding a European call for proposals is formally eligible and practically excluded. It is a vicious cycle.

This pattern generalises. Funding for FLOSS flows through a small set of channels: corporate sponsorship, Northern philanthropy, newer public instruments like Germany's Sovereign Tech Fund,11 and initiatives such as the Digital Public Goods Alliance that explicitly frame FLOSS as development infrastructure.12 Each is a genuine improvement on the scarcity that preceded it. Each also concentrates decisions about what deserves to exist in institutions where the periphery is not at the table. Priorities follow the funder's context: security audits for the software supply chains of Northern industry rank above localisation, accessibility, or the "boring" maintenance of tools that Southern civil society actually depends on.

The Balkans again make the mechanism visible. The region is close enough to the money to see it: its NGOs attend the same conferences, speak the grant dialects, sometimes win the grants after lots of stress and unpaid efforts upfront. And it is peripheral enough that the funding rarely builds anything durable. The BIRN findings cited earlier describe the result: an archipelago of platforms and projects that flourish for the length of a grant cycle and then decay.13 The lesson is not that funding fails, but that funding projects while ignoring communities and institutions fails. A grant can buy a codebase; it cannot buy the ongoing social structure that keeps a codebase alive, and it is precisely that structure that peripheral regions are not funded to build.

Bridges that exist, and bridges that are missing

"Bridging" is this report's assignment, so it is worth being precise about what a bridge is. A useful test: does the arrangement move power and capacity toward the periphery, or does it move the periphery's participation toward the centre? The first is a bridge. The second is often extraction wearing a lanyard. 

Real bridges exist, and most were built from the South rather than for it. Open Source Community Africa has grown from a meetup network into a continental infrastructure of chapters, mentorship and its own major conference, OSCAFest, a space where African contributors set the agenda rather than fill a diversity slot on someone else's.14 FOSSASIA has played a similar role across Asia for over a decade.15 Outreachy's paid internships attack the material barrier directly, on the correct theory that unpaid contribution is a regressive tax on the poor.16 Regional gatherings, from OSCAFest in Lagos to OSCAL in Tirana (now not active) to the FLOSSK conferences in Prishtina, do quietly radical work: they relocate the room where trust networks form, so that the periphery's contributors can build social capital without a Schengen visa.

Translation and localisation communities deserve particular recognition as bridge infrastructure. Every interface translated, every documentation set localised, every Wikipedia and OpenStreetMap contribution grown in a smaller language lowers the linguistic gate for thousands of people who will never appear in a contributor graph. That this work is chronically unfunded and treated as peripheral to "real" development says everything about whose contribution counts.

The hollow bridges are equally instructive. The diversity grant with no second year. The Northern conference that flies in a Southern keynote speaker but locates its programme committee entirely in the North. The corporate "open source for good" initiative that harvests goodwill and case studies, then sunsets when the marketing budget rotates. The mentorship programme that prepares peripheral contributors to fit the centre's norms while leaving the norms untouched. These fail the test above: they widen participation while conserving power. Communities on the receiving end learn to recognise them quickly, and the cynicism they leave behind is itself a barrier: the next, better designed initiative inherits the distrust.

What is missing, then, is not goodwill but architecture, and four pieces of it seem most urgent. The first is regional intermediaries. Funders cannot individually know 10,000 peripheral projects, but they can fund and trust regional organisations that do – an OSCA, a FOSSASIA, a Balkan hackerspace network – and devolve regranting decisions to them; intermediaries convert the periphery's illegibility from a disqualification into someone's job. The second is funding for communities, not only code: grants that can pay for a venue lease, a few pizzas (true story), a community manager, an accountant, childcare at a sprint (rarely), the unglamorous substrate of continuity that funding tied to projects structurally cannot buy. The third is governance quotas that bite. Foundations and programmes serious about the divide can measure it where it matters, in board seats, technical steering committees, programme committees and maintainer rosters, not in attendance statistics. The fourth is multilingualism as infrastructure: treating translation, localisation and plain language documentation as funded work of the first order, not volunteer overflow.

The open source “we” want

The divide this report has described will not be closed by the South "catching up", because it is not a gap in capability. It is a gradient of power that runs from an institutionally dense, well-funded core to a periphery that supplies a growing share of the movement's people and energy while holding a static share of its decisions. The Balkans' experience is a reminder that this periphery is not a hemisphere but a position, one that parts of Europe occupy alongside most of Africa, Asia and Latin America, and that marginalised communities occupy inside the North itself.

That reframing matters, because it changes what "bridging" means. If the divide were geographic, the answer would be outreach. Because it is structural, the answer is redistribution: of funding, of governance seats, of definitional power over what counts as contribution and whose problems count as core. The free software movement began as a political claim, that the people who use technology are entitled to control it. A movement with that founding claim cannot coherently maintain an internal arrangement in which most of its users, and an accelerating share of its builders, control so little of its direction.

The encouraging news is that the periphery is not waiting. It is building its own conferences, its own community infrastructure, its own norms, and on some measures of inclusion it is already ahead of the centre it is told to emulate. The open source we want would meet that momentum halfway: with funds that peripheral communities can actually reach, institutions they help govern, languages they speak, and a definition of merit generous enough to count the work that keeps communities, not just codebases, alive. The bridge, it turns out, is being built. It is being built from the far bank. The question this edition poses is whether the near bank will anchor its end.

 

Notes

This report was originally published as part of a larger compilation: “Global Information Society Watch 2026 special edition: Free/libre and open source software: Visions of openness from the Global South"

Creative Commons Attribution 4.0 International (CC BY 4.0) - Some rights reserved.

Web and e-book
ISBN 978-92-95113-91-6 
APC-202607-APC-R-EN-DIGITAL-384

Print
ISBN 978-92-95113-92-3
APC-202607-APC-R-EN-P-385

Themes
FLOSS
Free/libre open source software
Open source technology