---
title: "K-Net puts local-loop control where the repairs happen"
locale: "en"
url: "https://irz.fr/en/articles/knet-network-control-repairs-en"
markdown_url: "https://irz.fr/en/articles/knet-network-control-repairs-en.md"
category: "tech"
tags: ["networking", "telecommunications", "First Nations", "maintenance", "infrastructure"]
published_at: "2026-08-23T18:04:00.000Z"
author: "Hugo Marchal"
translation: "https://irz.fr/fr/articles/knet-controle-reparation-communautes-fr.md"
---

# K-Net puts local-loop control where the repairs happen

In northwestern Ontario, K-Net separates local ownership, regional aggregation and long-distance transport. Its durable innovation is less about the advertised speed than who can intervene when a connection fails.

K-Net's director described a very ordinary problem in 2022: when equipment fails in a remote community, bringing in a technician can mean putting that technician on a plane. K-Net changes the first step instead, by training somebody locally to look after the local loop while a regional service remains available when the fault grows beyond that first intervention; in the same year, 26 of the 100 Ontario First Nations connected through the wider K-Net system were operating community ISPs of their own.[4](https://nationalpost.com/feature/left-behind-indigenous-communities-internet)

That division of labour makes a worse poster than a speed record, but it helps explain why the Kuhkenah Network, created by the Keewaytinook Okimakanak tribal council in northwestern Ontario, is still active more than thirty years after its first mandate. K-Net is not one cable owned end to end by the participating communities; it is an arrangement in layers, with local networks managed in each First Nation, regional aggregation and shared services, then public or commercial long-distance infrastructure carrying traffic towards the rest of the internet.

## A floppy disk by plane

Keewaytinook Okimakanak asked its computer service to close a communications gap in 1994, and K-Net began with a “stay in school” bulletin board that let young people who had moved to Sioux Lookout for secondary school exchange messages with relatives at home.[1](https://utoronto.scholaris.ca/items/8c1564b2-7a5c-4c72-b65c-42d27fa33838)[2](https://kochiefs.ca/k-net/)

The connection was hardly seamless. In 1995 and 1996, schools used the telephone network, paid long-distance charges and reached at most 14.4 kbps; some sites that remained offline periodically sent floppy disks by plane to the server in Sioux Lookout and received mail the same way.[1](https://utoronto.scholaris.ca/items/8c1564b2-7a5c-4c72-b65c-42d27fa33838) Before there was a broadband problem, there was simply the problem of finding a path that existed.

Broadband arrived in 2001 through the SMART demonstration project. Bell supplied 1.544 Mbps T1 lines to five communities, while Fort Severn, too remote for the same connection, used satellite; federal and provincial programmes funded the stage, and later agreements with Telesat and other public initiatives extended the network.[1](https://utoronto.scholaris.ca/items/8c1564b2-7a5c-4c72-b65c-42d27fa33838)

> K-NET
> **Thirty years of network**
> - A BBS links students and families.: 1994
> - T1 and satellite bring broadband.: 2001
> - The NWOBEI loop enters its upgrade project.: 2023
> - 18 First Nations announce completion.: 2026
> Sources: CRACIN, K-Net.

Those names matter. Describing the project as if First Nations built every kilometre alone would erase Bell, Telesat and millions in public funding, while describing the communities as passive recipients would erase the architecture choices, the local operators, the training and the social priorities that shaped how regional capacity was used.

## Three pieces

In the model documented in 2006, each participating First Nation owns and manages its local network, the *local loop*. The band office may provide internet service itself or hand the job to a community company, while managers and technicians, sometimes shared with a neighbouring community, handle repairs, upgrades and ordinary maintenance.[1](https://utoronto.scholaris.ca/items/8c1564b2-7a5c-4c72-b65c-42d27fa33838)

K-Net stays behind that local layer rather than replacing it. The regional team designs the wider architecture, buys bandwidth in aggregate, operates support and keeps expertise that would be expensive to reproduce in every small community; technicians also exchange solutions with one another, so local control does not mean leaving one person alone in front of a dead network cabinet.

The long-haul links form another layer in the 2006 account: Bell fibre, Telesat capacity, urban points of presence and interconnections with other carriers. No band council owns every metre between a northern home and Toronto, but the participating community controls the part containing its subscribers, its local equipment and the people who repair it, which is a narrower and more credible meaning of “community-owned internet”.

## Bandwidth has an order

Aggregation was also a way to manage scarcity. On the older satellite network, cutting capacity into fixed slices would have prevented a community from temporarily borrowing enough bandwidth for a telehealth call, so K-Net kept capacity in a common pool; when a medical videoconference needed more, ordinary traffic elsewhere could slow without another community being pushed off the network entirely.[1](https://utoronto.scholaris.ca/items/8c1564b2-7a5c-4c72-b65c-42d27fa33838)

A public event in 2007 connected participants from twenty-two communities or sites in one videoconference. More generally, limited satellite capacity meant sessions had to be booked, and scheduled traffic for health, education or collective meetings could receive dynamic priority.[3](https://publications-cnrc.canada.ca/eng/view/accepted/?id=c59c59ca-7a6d-4683-98fe-0565ad2a38af)

That makes “control” quite practical. Owning a router is less important than being able to decide that a medical examination or a class should move before somebody else's download, and the *National Post* reported in 2022 that K-Net could still manage traffic to prioritise the telehealth and education services carried by the network.[4](https://nationalpost.com/feature/left-behind-indigenous-communities-internet)

## Ten gigabits where?

The upgrade started in 2023 on the NWOBEI fibre loop did not replace this federated architecture. Eighteen First Nations were included, and the project combined local infrastructure, K-Net systems and Bell fibre with new 10 Gbps point-of-presence equipment, wireless access, backup batteries and training for community technicians.[5](https://knet.ca/knet-announcement-for-knet-upgrade-to-increase-bandwidth-for-18-first-nations-in-northwestern-ontario-on-the-nwobei-fibre-loop/)

K-Net announced completion in January 2026. According to the organisation, homes inside the covered area can now obtain at least 50 Mbps downstream and 10 Mbps upstream, while head-end equipment can reach 10 Gbps.[6](https://knet.ca/broadband-upgrade-project-complete-in-18-first-nations/) The two figures live at different points in the system: 50/10 describes the residential availability K-Net advertises, whereas 10 Gbps describes equipment at a point where many connections meet.

The extra capacity was needed, but higher numbers do not remove the maintenance problem that shaped K-Net in the first place. Fibre ages, electronics fail and one urgent use will eventually compete with another, so the responsibilities remain split: the community maintains its local network and decides how service is provided, K-Net aggregates the work that benefits from scale, and carriers move traffic farther away.

The floppy disk no longer has to board an aircraft. The more durable part of the design is that the first person able to repair the connection may already be in the community when it breaks.

## References

1. [Adam Fiser, Andrew Clement and Brian Walmark, The K-Net Development Process: A Model for First Nations Broadband Community Networks, CRACIN, 2006](https://utoronto.scholaris.ca/items/8c1564b2-7a5c-4c72-b65c-42d27fa33838)
2. [Keewaytinook Okimakanak, K-Net](https://kochiefs.ca/k-net/)
3. [Fenwick McKelvey and Susan O’Donnell, Multi-site Videoconferencing as a Public Sphere in First Nation Communities: A Case Study, NRC, 2009](https://publications-cnrc.canada.ca/eng/view/accepted/?id=c59c59ca-7a6d-4683-98fe-0565ad2a38af)
4. [Jessica Mundie, Many Indigenous communities lack internet infrastructure. Some are building it themselves, National Post, 29 December 2022](https://nationalpost.com/feature/left-behind-indigenous-communities-internet)
5. [K-Net, KNET Announcement for KNet Upgrade to Increase Bandwidth for 18 First Nations, 27 July 2023](https://knet.ca/knet-announcement-for-knet-upgrade-to-increase-bandwidth-for-18-first-nations-in-northwestern-ontario-on-the-nwobei-fibre-loop/)
6. [K-Net, Broadband Upgrade Project Complete in 18 First Nations, 13 January 2026](https://knet.ca/broadband-upgrade-project-complete-in-18-first-nations/)