Summary

  • UbuntuNet is a non-profit membership-governed organisation registered in Malawi, serving national research and education networks in East and Southern Africa; its records use the descriptors "association" and "Trust".
  • It has aggregated regional capacity through carrier contracts, points of presence, member routes and AfricaConnect programmes, including 10 Gbps core links and larger national connections.
  • Its portfolio has moved beyond transit to identity, cloud, storage, open science, cybersecurity and NREN development, with varying service maturity.
  • Aggregated demand and expertise create value, but weak university links, arrears, supplier power, conflict, grant cycles and limited disclosure can prevent regional capacity from reaching researchers.

Network 3 upgraded the core to a 10 Gbps operational tier

The 2023 report describes the transition of core links from around 2.5 Gbps to 10 Gbps. It also describes at least 20 Gbps of total capacity through redundancy. These two figures represent different levels and do not imply a uniform 20 Gbps across every path.

The most consistent reading is that there are backup 10 Gbps core links on certain segments, with speed varying by route, carrier and national member connection.

This represents a significant advance in price and flexibility, even if it remains below networks of hundreds of gigabits or terabits in wealthier regions. Any comparison should explain the region’s economics and starting point, rather than reducing speed to a simple ranking.

The drop in unit price was a structural outcome, not a side effect

The 2023 report noted that unit prices fell by between 50% and 71.5% compared with 2022. Some networks moved from services between roughly 130 Mbps and 4 Gbps to 500 Mbps–8 Gbps, depending on their circumstances.

The reduction came from collective purchasing, long-term contracts and better core utilisation. It proves that the aim is not merely to announce a higher capacity figure, but to deliver more capability per unit of budget.

However, passing the savings on to members leaves the alliance with a thinner financial margin. Currency swings, a single invoice or a delayed payment can eat into it. The lower price stays sustainable only if recurrent costs are understood and paid on time.

BotsREN and ZIMREN show how regional upgrades reach national networks

BotsREN connected at 10 Gbps in 2023 and then upgraded to 20 Gbps in January 2026. ZIMREN connected at 10 Gbps in June 2025. These are national access speeds, not the speed of the entire network.

Nevertheless, they demonstrate that core investment translates into a measurable service for the member. The ultimate value depends on how many institutions sit behind the NREN, the local loops, and the equipment and skills inside the universities.

A 20 Gbps national link can terminate at a university with only 1 Gbps, unstable power or an overloaded firewall. Connection figures should therefore be accompanied by usage, loss, availability and coverage data.

The regional layer begins where national networks end

A student or researcher rarely sees the name UbuntuNet Alliance in the service they use. The journey typically starts with a university or research institution network, then moves to a national research and education network such as KENET, RENU, TENET, ZAMREN, TERNET or MoRENet. UbuntuNet becomes important at the next level, when the national network needs to exchange traffic with another country, reach an international scientific facility, connect to GÉANT, or use a shared service that one country would struggle to afford alone.

This position explains why the alliance is critical without looking like a consumer internet provider. It does not usually sell home connections, does not connect every university directly, and does not run campus Wi‑Fi. It operates a regional layer of routes and services on top of independent national systems. Other regional networks, telecoms operators, commercial clouds, exchange points and the receiving institution complete the rest of the path.

The user perceives one session, but operational authority is distributed. UbuntuNet can adjust its routes, increase a core link’s capacity, monitor a regional service and coordinate incident response. It cannot restore power to a university, fix an outdated institutional directory, force a local carrier to repair a cable, or guarantee the capacity of a remote application. Its task, therefore, is not to suggest a single owner, but to make independent domains work together within understood responsibility boundaries.

The federation is the design itself, not a temporary stage before central ownership

UbuntuNet Alliance is a federation of NRENs, not a holding company. Each member network keeps its legal identity, staff, budget, local infrastructure, university and government relationships, and pricing policy. The regional layer exists because the members choose to buy some capacity together, exchange traffic at shared locations, run collective services, and represent their interests regionally and globally.

This model suits a region where telecoms markets, currencies, higher-education systems, procurement rules, languages and technical maturity levels differ. Building a single centrally owned academic network across all countries would require political authority that UbuntuNet does not possess – nor needs. A federation instead combines national independence with regional scale, and lets more mature networks transfer expertise to emerging ones.

Complexity does not disappear; it shifts to the handover points. A slow file transfer can be affected by a university firewall, a national circuit, a regional point of presence, an international carrier or the receiving system. Each party sees only part of the evidence. This is why service boundaries, shared measurement, escalation channels, change procedures, contribution rules and workable exit options become part of the infrastructure.

The legal record is operationally valid, but the public terminology is not uniform

Current materials describe UbuntuNet as a non-profit association registered in Malawi. The 2023 annual report and historical transition documents also use the descriptor of a non-profit Trust governed by members and trustees. The current constitutional document was not available in the research pack, so the discrepancy should not be erased by assuming one description is wrong.

The practical structure is clearer. UbuntuNet has members, not ordinary shareholders. Documented governance consists of a members’ assembly, a board of trustees, and a secretariat led by a CEO. The organisation can employ staff, receive grants, sign contracts, own or lease equipment, and operate services. There is no evidence of distributable ownership or a conventional commercial valuation.

Calling it a “private telecommunications company” suggests investor control and profit distribution. Calling it only a “voluntary community” understates its contractual and fiduciary obligations. The safest formulation is: a non-profit, membership-based organisation registered in Malawi for NRENs, noting that its sources use the words association and Trust.

The foundational problem was economic and geographic, not a complete absence of the internet

The region’s universities were already connected to the internet before UbuntuNet was founded. What was missing was a permanent regional mechanism that could aggregate academic demand, exchange research traffic, and link to global research and education networks on terms that suited science and education. Institutions were buying small amounts of international capacity at high cost, and traffic between two African countries sometimes travelled via Europe.

Commercial connectivity provides general access, but it does not automatically create direct paths between neighbouring NRENs, predictable capacity for data-intensive science, federated identity, shared security practices, or collective bargaining power. Small networks were weak in the face of suppliers, while a new NREN struggled to demonstrate its value before it had enough members and capacity.

The alliance changed the unit of demand. Universities’ requirements are aggregated at the national level, and then the national networks’ requirements are aggregated regionally. A larger collective buyer can plan points of presence, sign long-term contracts, build operations and negotiate with global partners. UbuntuNet did not abolish the commercial market; it entered it as a coordinated buyer and a mission-driven operator.

Five NREN initiatives formed the initial coalition

The UbuntuNet Alliance was conceived during the second half of 2005 with the involvement of initiatives in Kenya, Malawi, Mozambique, Rwanda and South Africa. The founding base is linked to KENET, MAREN, MoRENet, RwEdNet and TENET, as those entities existed at the time. They were not all at the same level of maturity.

TENET and KENET brought operational experience and an established university base. MAREN provided a foothold in the country that would later host the secretariat. MoRENet represented a government-backed national model. RwEdNet later demonstrated both promise and fragility; its collapse and rebuilding showed that a regional layer cannot forever substitute for a sustainable national institution.

The diversity was a strength. Had the alliance been formed only of mature networks, it might have become a limited interconnect club. Had it consisted only of weak initiatives, it would have lacked traffic, expertise and credibility. The coalition brought together the capacity to run a real service and the task of building national capabilities.

Francis Tusubira helped turn a network proposal into an institution

Francis “Tusu” Tusubira is associated with both the founding phase and the wider African research and education networking movement. The task was not simply to choose equipment and routes. Universities, ministries, regulators, telecoms operators, donors and engineers all had to be convinced that a regional research network should be a permanent institution, not a project that ends when funding stops.

That required legal personality, membership rules, accounts, contracting capacity, an annual forum, and a shared understanding that an NREN is more than a cheap internet provider. Tusubira’s contribution should be placed inside this collective construction, without erasing the roles of the founding networks, boards, technical communities and international partners.

The durable output was an organisation that could receive programme funds, buy capacity over the long term, and represent a regional community. That institutional capacity enabled the subsequent AfricaConnect phases.

Amsterdam provided the first legal vehicle; Malawi became the regional seat

Registration documents were filed in Amsterdam in 2006. For an organisation that was still taking shape, the Dutch entity offered a familiar framework for European partners for contracts and projects before a stable legal base within the region could be completed.

As the alliance matured, a European legal centre became less aligned with its mission. A new constitution was adopted in Malawi in 2012, registration followed in 2013, and historical documents describe the transfer of assets and obligations to the Malawian structure. The secretariat is now based in Lilongwe.

The move brought governance closer to the members but left the association–Trust inconsistency in place. An accurate narrative must distinguish the 2005 conception, the initial legal encapsulation in the Netherlands in 2006, and the subsequent legal and administrative migration to Malawi during 2012–2013.

Early transit to GÉANT proved the value before the backbone was complete

Historical materials point to the implementation of a research and education transit service towards GÉANT in 2008. However modest, it gave members a path to European and global research networks and demonstrated that a regional institution could deliver value before a mature regional network had been completed.

The relationship was not limited to capacity; it also encompassed procurement know-how, programme management, identity, security, operations and access to European funding programmes. The partnership accelerated construction but also linked part of the expansion to European programme cycles and priorities.

It is inaccurate to describe the infrastructure as wholly imported, or to describe UbuntuNet as completely independent. Global partnerships were used to build regional capacity governed by the members, and the ongoing question is which services can be transitioned to business-as-usual operations after a project ends.

Long-term carrier contracts turned aggregated demand into infrastructure

The fifteen-year contracts with WIOCC in 2013 and SEACOM in 2016 gave the alliance a more stable horizon than small annual contracts. Long contracts can lower unit prices, guarantee specific routes, improve diversity and support upgrades.

But they also create long-term commitments. The alliance remains exposed to supplier performance, technology change, maintenance, foreign exchange and continued payments. An international carrier invoice can fall due even while a national network’s contribution is late.

These contracts reveal how a federation becomes infrastructure: an organisation aggregates demand, signs on behalf of several members, places equipment and coordinates repair. It does not, however, become the owner of every fibre, landing station or transit network used.

The 2014 go-live turned the alliance from a promise into a production operator

In July 2014, UbuntuNet Alliance and DANTE announced the activation of regional links at 622 Mbps. The speed looks small today, but it was a big step for networks that had previously bought much smaller capacities at high prices.

From that point on, real production traffic appeared, along with incidents, configurations, shift schedules and member expectations. A routing mistake was no longer an internal project issue; it was an incident affecting universities and laboratories.

That first go-live created points of presence, NOC operations, carrier relationships and billing mechanisms that later phases could scale, rather than having to start from scratch.

AfricaConnect2 expanded the programme and strengthened eastern path resilience

AfricaConnect2 was announced in 2015 as a €26.6 million programme covering several regions and partners, among them UbuntuNet, WACREN, ASREN and GÉANT. That figure is not a standalone revenue line for the alliance.

For UbuntuNet, the phase expanded routes and membership, and underpinned the SEACOM contract connecting Kampala, Dar es Salaam and Amsterdam. That path added diversity in the east and reduced dependence on a single route.

The phase also reinforced the federal African model: RRENs retain their governance and build their own systems within a shared framework. Outcomes remained uneven because markets and national institutional structures differ.

AfricaConnect3 made the service portfolio as important as the backbone

AfricaConnect3, whose total programme reached €37.5 million and ended in April 2025, supported the network, pricing, training, identity, cloud and open science. Success could no longer be measured only by the number of circuits, but by members’ ability to use the shared infrastructure.

The phase produced Network 3, new points of presence, faster national connections and lower prices, and also expanded federated identity, repositories, Utafiti Africa, AfricArXiv and cybersecurity.

The expansion is logical because fibre without identity, storage or staff can remain unused. But it increases the maintenance burden; each added service brings software, data, security responsibilities and recurring cost.

AfricaConnect4 opens a new investment window and sets a deadline for sustainability

AfricaConnect4 began in 2026 with a further €40 million of European support over four years, channelled through the regional research networks in sub-Saharan Africa. The figure is a programme-level amount, not a grant earmarked exclusively for UbuntuNet.

The phase includes national and campus layers, cybersecurity, weather stations and climate data, EUMETCast, federated computing, GPU units and open science. It recognises that a fast backbone is not enough if institutions cannot use it.

By 2030, the weather stations, repositories, identity, cloud and computing may remain while the initial funding ends. The permanent operator, budget, fee model, staffing and conditions for termination or migration must therefore be defined for each service.

AS36944 defines UbuntuNet’s public boundary in routing

UbuntuNet uses autonomous system AS36944. PeeringDB and BGP observers confirm the identity, some peers, prefixes and RPKI validity for observed assets. They reveal the public edge, not all private circuits.

PeeringDB shows a self-declared traffic range of between 5 and 10 Gbps, along with around 1,200 IPv4 prefixes and 150 IPv6 prefixes. bgp.tools observes a smaller set of directly originated assets. The difference is normal because each source measures a different scope.

These metrics do not proxy for core capacity or member connectivity. The ASN confirms operational BGP participation, but does not prove a particular level of availability, peak traffic or physical diversity.

Points of presence and exchanges create options, but the map is not an asset register

The 2023 report lists locations in Cape Town, Mtunzini, Maputo, Dar es Salaam, Nairobi, Amsterdam, Gaborone and Johannesburg, describing Djibouti and Mombasa as planned or under development. Older pages mention London and other sites.

The differences may reflect different dates, or definitions of what counts as a point of presence, a handoff or a routing change. A dot on a map does not reveal who owns the router, fibre, rack, port or contract.

Johannesburg and NAPAfrica added southern options. But real resilience depends on the independence of ducts, providers, power, data centres and software. Two mapped paths can still share a single physical point of failure.

The Lagos–Cape Town path made African federation more tangible

Developments in 2025–2026 connected WACREN to the South African environment through ZAOXI, SANReN and TENET, and then to UbuntuNet. Partners described it as the first operational direct interconnection between African regional research networks on African soil without transiting Europe.

The statement must remain attributed and scoped to the RREN category. It does not mean no earlier commercial or private links existed within the continent; it points to the institutional and operational relationship between regional research networks.

The path can keep more scientific traffic inside Africa and reduce some detours. It still depends on carriers, exchange points, BGP policies and national partners. No single party controls the entire route.

Membership numbers reflect different categories and dates

The 2023 report recorded fifteen active members, more than 3.5 million users, and over a thousand institutions in 13 member countries. A March 2026 due-diligence request listed networks from 18 countries, while the website displayed 16 profiles. The numbers may measure legal membership, activity, connectivity or website freshness.

Lesotho was welcomed in November 2025, and trustee elections took place in 2026. Every number should therefore be accompanied by a date and a definition.

Membership does not prove a production connection, and a country connection does not mean uniform service to every university. In conflict-affected areas, legal status may diverge from actual operations.

Campus infrastructure is the decisive test of regional investment

A fast core cannot repair poor local fibre, ageing Wi‑Fi, frequent power cuts, overloaded security appliances or staff shortages. The weakest link determines the researcher’s experience.

That is why AfricaConnect4 includes national and campus components. Regional capacity can remain unused if institutions lack storage, servers, transfer tools and adequate internal networks.

Responsibility is distributed: UbuntuNet coordinates the regional layer, the NREN builds the national network, the university runs the campus, and governments, carriers and donors set the environment. Impact should be measured down to the researcher, not stopped at a dot on a country map.

Lilongwe hosts governance; network operations concentrate in Kampala

The secretariat is based in Lilongwe, Malawi, while organisational materials place the network operations centre in Kampala, Uganda. That befits a distributed regional organisation.

The secretariat manages members, contracts, projects and finance. The NOC monitors links, opens tickets, coordinates with suppliers and supports NRENs. The model requires secure access, clear authority boundaries and continuity plans.

It also demands documentation, shift knowledge, credential storage and succession. A small organisation cannot afford to have understanding of a critical system depend on a single person.

Shared operations should strengthen national capability, not replace it

Not every NREN can run a 24/7 operations centre. UbuntuNet can offer monitoring, preventive maintenance and first- or second-level support, while the national network retains its customers, policies, suppliers and local authority.

The service succeeds when it fills a gap and transfers knowledge. It becomes a risk if the member no longer understands its own configuration or cannot leave. Change rights, maintenance windows, logs, escalation, costs and a handover plan must all be specified.

The regional centre can diagnose and coordinate, but it cannot repair a local cable without the carrier. Its value lies in linking evidence and preserving shared operational memory.

Federated identity is a second network built on institutional trust

The physical network moves packets; a federation lets a service trust a user authenticated by another institution. The university manages the account, a national federation coordinates rules and metadata, and regional and global layers connect the circles of trust.

UbuntuNet does not centrally store every user’s password. It operates or supports proxies, federations, metadata and enrolment into eduGAIN. Quality remains hostage to institutional directories, the account lifecycle, MFA, certificates and attributes.

This enables an institution’s identity to be used for libraries, clouds, repositories and computing. It also means compromising one identity provider becomes a cross-service risk. Federated trust must therefore be accompanied by federated audits and revocation procedures.

eduroam turns campus identity into a mobility service

eduroam lets a student or researcher use their home institution’s credentials at a visited university. An authentication request may traverse local, national, regional and global RADIUS proxies before returning to the home institution.

The seamless experience depends on certificates, realms, secure Wi‑Fi and accurate accounts. UbuntuNet can run the regional layer, but it cannot fix an expired account or a poorly configured access point.

Success should be measured by active institutions, successful authentications, incidents and resolution time – not by the count of installed servers alone.

eduID.africa and eduGAIN open a path to wider trust systems

eduID.africa works with WACREN and ASREN as a hosting federation and a model for networks that have not yet built a mature national federation. It allows participation before each country rebuilds the full stack.

MAREN and TERNET joined eduGAIN in 2025, linking the Malawian and Tanzanian federations to the global federation. Joining proves institutional compatibility, but it does not prove that every university has equivalent directory quality, attributes and assurance.

The regional layer should accelerate national capacity building and provide a clear transition. If it becomes a permanent, opaque substitute that is hard to leave, it weakens independence.

Research4Life turns federation into practical knowledge access

A project launched in 2025 allows institutions in Kenya, Malawi, Uganda and Zambia to access Research4Life resources remotely with their institutional credentials. A later update reported that 66 institutions in 15 countries had joined.

This reduces reliance on shared passwords that are hard to revoke or audit. Access can be tied to the individual and their affiliation, without creating a separate account with each resource.

The figure is date-stamped and does not imply universal coverage for every member. Institutions remain responsible for identity, and Research4Life for eligibility terms. Still, the example shows tangible value of federated infrastructure for the researcher.

Cloud services combine regional infrastructure with commercial procurement

One branch is infrastructure hosted or managed by the community: UbuntuNet Open Science Cloud, file transfer, Drive, hosting, collaboration and experimental computing. Another branch makes commercial services available, most notably the Redington/AWS agreement in 2024.

A regional cloud can keep skills and decisions closer to the members, while a hyperscale cloud provider offers a product range a small RREN cannot replicate. UbuntuNet aggregates demand and support, but it does not own AWS or control its pricing, availability or product roadmap.

The 2023 report documents an installation in Lusaka and equipment for Kigali, but does not publish a uniform capacity, utilisation figure, tenant count or current SLA. Installed hardware does not become a mature multi-site cloud without orchestration, identity, backups, security, support and funding.

Research support services address everyday friction alongside big-science needs

Utafiti Africa aggregates funding and training opportunities and passed 1,100 active users in 2024. It does not award funds; it lowers the cost of finding opportunities.

File transfer, Drive, Meet, Learn and hosting can reduce duplicate purchases and ease collaboration. But the maturity of these items is uneven, and some older pages mix a running service with a historical plan.

Moving a large file once is not the same as running steady streams from telescopes or climate models. Heavy workloads require tuned end points, storage, monitoring and possibly dedicated circuits. The portfolio becomes more reliable when each service has a clear scope.

Open science became a second layer of infrastructure

A network can transport a paper or a dataset, but it does not automatically make it discoverable, citable or preserved. Repositories, metadata, identifiers and staff also depend on hosting, storage, identity, certificates, backups and durable institutions.

UbuntuNet and Access 2 Perspectives have hosted the AfricArXiv infrastructure since late 2023. AfricArXiv remains a community-led initiative, not owned solely by UbuntuNet. The alliance was also selected as the Southern Node of the Africa Open Science Platform.

Repositories have a long lifecycle. Software, rights, records, security, preservation and data export must all be maintained if the partnership changes. Abandoning a scientific record may be more damaging than replacing a circuit.

Repository stewardship and persistent identifiers demand sustained institutional work

The Research Repository Stewardship Programme trains librarians and data managers. The first Portuguese-language workshop was held in Mozambique in July 2026, showing that localisation is part of operations, not a communications ornament.

Institutions need deposit policies, metadata, rights, support, backups, upgrades, preservation and a budget. Multi-tenant platforms reduce duplication but create a shared failure domain. The Network Adoption Fund aims to turn NRENs into sustainable service providers.

DataCite, DOIs, ORCID and the 2026 DOCiD partnership with TCC Africa add an identification layer for people, institutions and outputs. An identifier cannot fix poor metadata. UbuntuNet can coordinate procurement and integration while keeping global governance and local accountability clear.

Cybersecurity capability is distributed across individuals and institutions

Security work includes regional camps, national workshops, identity and systems training, and participation in the R&E security community. There is no evidence of a single security operations centre monitoring everyone. The more accurate model is to strengthen national and institutional teams and build shared contacts and procedures.

Laws, data and incident authority remain national or institutional. A regional organisation can share tools and knowledge, but it does not automatically remediate a compromised machine inside a university.

Training for finance officers from 15 countries in 2026 shows that resilience also covers budgets, currencies, compliance and collecting contributions. If a capacity contract or a grant cannot be renewed, technology alone does not make the structure resilient.

Governance maintains continuity despite gaps in public information

The members’ assembly is the supreme authority, the board of trustees oversees, and the secretariat executes. Professor Madara Ogot was appointed CEO for a four-year term from 1 February 2022; 2026 materials still refer to him as CEO, but no separate renewal announcement was found. Professor Hellicy Ng’ambi became the fourth Chairperson on 1 January 2024.

Miriam Chahuruva and Sabelo Dlamini were elected as trustees on 29 May 2026. Post-election sources do not publish a fully identical list or the names of all departing trustees.

This does not prove a vacuum, but it limits the precision that can be published and underlines the importance of keeping the governance record publicly current.

The 2023 accounts reveal a large mission and a thin financial reserve

UbuntuNet reported income of $2,404,779, expenditure of $2,393,911 and a surplus of $10,868. Connectivity income was $1,902,318, or 79.1%, while network operating costs were $928,032.

The margin is small relative to the mission and risks. A currency move, an invoice or a payment delay can consume it. Passing purchasing savings on to members fulfills the purpose but thins the reserve.

The CEO noted that late operations and maintenance contributions from NRENs severely affect the ability to meet obligations. No fully audited accounts for a year after 2023 were found, so the current financial position cannot be inferred from the AfricaConnect amounts.

Donors, carriers, currencies and conflict stay inside the operating model

AfricaConnect funds major expansion, commercial carriers supply a large part of the physical infrastructure, clouds control their own platforms, contracts are paid in hard currency while NRENs receive local resources, and conflicts can reduce member capacity.

RwEdNet collapsed in 2018 and was rebuilt in 2024 – direct evidence that a national institution can be fragile even when a regional backbone exists. Membership of a conflict-affected state also does not guarantee that service continues.

These dependencies do not nullify the alliance’s usefulness; they define the work: diversify suppliers, build reserves, strengthen national capacity, document responsibilities, and understand sovereignty as choices, not isolation.

Public evidence is strong on programmes and weaker on real-time operations

UbuntuNet publishes a great deal about contracts, links, elections, training and service launches. That is enough to reconstruct the history. But there is no current list of every route and asset, no multi-year availability record, no full traffic series, no per-link utilisation, no public incident database, no current staff headcount, and no accounts after 2023.

PeeringDB, bgp.tools, core capacity and member speed all measure different things. A point of presence can mean a router, a handoff or a facility presence that does not reveal physical ownership. An independent SLA cannot therefore be calculated from these sources.

Not publishing an incident does not mean none occurred, and a lack of metrics does not automatically mean performance is poor. Publishing availability, restoration, utilisation and service adoption would improve accountability to members and donors.

Regional autonomy comes from multiple options, not from eliminating external dependence

UbuntuNet increases regional control through collective purchasing, keeping more traffic inside Africa, training engineers, and operating identity and repositories. It does not eliminate global standards, commercial cables, hardware vendors, hyperscale clouds or scientific partners.

The relationships with WACREN, ASREN, SANReN, TENET and GÉANT illustrate an African federation that no single party owns. The Lagos–Cape Town path improves traffic locality, but it depends on carriers, exchange points and shared policies.

Every choice has a trade-off: an AWS framework eases procurement and deepens platform dependence; a regional repository strengthens governance and needs software and staff; a long contract lowers price and increases supplier and currency risk.

The strategic test is the ability to change. Can a member switch carrier, export data, move workloads and keep its identity? Autonomy grows when the answer is yes and exit is practical.

UbuntuNet turns separate networks into a usable research environment

The fundamental achievement is both institutional and technical. Independent NRENs aggregate demand, share operations, connect globally, and build services that would be costly or absent if each country acted alone. The backbone, identity, repositories, cloud and training are different layers for solving the fragmentation problem.

The organisation remains incomplete. Its capacity is limited relative to climate, astronomy and AI ambitions; public information is stronger on achievements than on usage and availability; member funding is uneven; and suppliers, donors and services of varying maturity remain inside the system.

These limitations do not diminish its importance. UbuntuNet is not a “separate African internet” nor a finished continental platform. It is the shared regional layer that makes a wider system possible. Success will be measured by whether members can use, fund and monitor the services, and move data, identity and themselves out when needed without losing their independence.