WOPR Systems — Community-owned mesh networks and sovereign tech, not Big Tech.

Community Owned Internet Explained

The 10-second version

Community owned internet means the people who use a network also own and govern it, rather than buying access from a distant company.

It takes four common forms: municipal fibre, member-owned co-operative ISPs, volunteer wireless mesh, and community-hosted services like a shared Nextcloud.

It is more accountable and more durable over time, but you trade corporate convenience for local labour and slower support.

Community owned internet is network infrastructure that the people using it also own and govern, instead of renting it from a company headquartered somewhere else. The fibre, the rooftop radios and the servers belong to a co-operative, a nonprofit, a town, or a neighbourhood association. Members pay in, decisions get made in a meeting anyone can attend, and money left over after costs goes back into the network rather than out to shareholders.

A useful comparison is the difference between renting an apartment and joining a housing co-op. Either way you are connected to the same city water main. But in one case you hold the keys, you vote on the roof repair, and nobody raises your rent from a boardroom you have never seen. Almost every community network still buys upstream capacity from a larger carrier, but the part that touches your street is locally held and locally answerable.

What Community Owned Internet Actually Means

The phrase gets used loosely, so it helps to have a test. Three questions separate a genuinely community owned network from a company with friendly branding.

Notice what is missing from that list: the technology. A community network can be fibre in a trench, radios bolted to chimneys, or refurbished servers in a library basement. Ownership is a legal and governance question, not a hardware question. Two neighbourhoods can run identical equipment and only one of them owns its internet.

It does not mean free, either. Somebody still pays for upstream capacity, power and replacement gear. What changes is who that somebody answers to, and whether a bad decision gets you a support ticket or a vote. It is data sovereignty applied to wires.

The Four Models You Will Actually Encounter

Projects that call themselves community owned almost always take one of four shapes. They are not exclusive, and mature projects often run two or three at once.

That last model is the cheapest door to walk through and the one most groups should try first. It needs no permits and no trenching, and it builds the thing every network depends on: a few people who have run something together and are still speaking to each other.

Who Owns Which Layer

Nobody owns the internet, and no community network owns the whole path between your laptop and a server in another country. The honest way to look at it is layer by layer, asking which parts are realistically within reach of a group of neighbours.

LayerWhat it isWithin community reach?
Apps and servicesFiles, chat, mail, video callsYes, the easiest layer to own
Last mileThe fibre drop or rooftop radio to your buildingYes, and the usual target
Middle mileLinks between neighbourhoods, dark fibre, tower spaceSometimes owned, often leased
Internet transitBulk capacity to the rest of the worldRarely owned, almost always bought
Exchange pointsWhere networks hand traffic to each otherShared, through membership
Poles, conduit, spectrumThe right to put equipment somewhereNo, granted or rented

Most successful projects own the middle of that table and buy the top and bottom. That is not a failure of ambition. It is how every network works, including the large carriers, who buy transit from each other constantly.

The strategic point is that the layers you can own are also the layers where the surveillance and the price gouging happen. Owning your last mile and your services removes the third party sitting between you and the people you talk to. Whether the packets cross an ocean on leased capacity matters far less than whether somebody is reading them on the way out of your building.

How The Money And Governance Work

Community networks fail for boring reasons far more often than technical ones. The two that matter are money structure and governance, and both need settling before the first radio goes up.

Separate capital cost from operating cost. Capital is one-time: radios, switches, a server, conduit, a ladder. Operating is what recurs whether anyone uses the network or not, chiefly upstream transit, power, renewals, and a replacement fund for hardware that will eventually die. Groups routinely raise capital through grants or member buy-in, then collapse two years later because nobody budgeted the replacement fund. A network with no maintenance money has an expiry date.

Write the bylaws while the group still likes each other. Decide how members join and leave, who can spend money and up to what limit, how disputes get settled, and what happens to the equipment if the project winds down. One member one vote is the co-operative standard, worth keeping even when one person does most of the work. Once a group outgrows a show of hands, verifiable voting software such as Folkmoot lets members confirm their own vote was counted without trusting whoever ran the count.

Volunteer labour is the other quiet subsidy. It is real and it is not free. A network that depends entirely on one tireless person has a bus factor of one.

The Software And Hardware That Does This

None of this requires inventing anything. The tooling for community networks is mature, open source, and mostly boring in the good way.

The hardware end is similarly unexotic: radios in unlicensed bands, a mast, a grounding kit, weatherproof cable. The expensive part is rarely the electronics. It is roof access, mounting, and the person willing to be up there in February.

What Community Owned Internet Is Genuinely Good At

Three things make this worth the trouble, and it is worth being specific rather than gesturing at freedom.

It reaches places carriers skip. A commercial ISP builds where the return justifies the trench, which leaves rural valleys, small towns and low-income blocks unserved for decades. Not because the engineering is hard, but because the spreadsheet says no. A co-operative answers to residents rather than a margin target, so the same street becomes worth serving.

It keeps working when other things stop. A local mesh with local servers still carries messages between neighbours when the upstream link is cut, because that traffic never needed to leave. During storms and outages, that is the difference between a neighbourhood that can coordinate and one that cannot. A carrier outage takes every service with it.

It removes the incentive to watch you. Traffic on a network your community owns is not an asset to be enriched and sold, because no shareholder is asking why the data is not being monetised. Run your own services on top and the party in the middle disappears entirely, which is the same reasoning behind moving off Big Tech services.

The Honest Limits And Tradeoffs

This is where most writing on the subject goes quiet, which is exactly why community projects start on false expectations and then stall. Here is what community owned internet is not good at.

If you want something that works the moment the technician leaves and never asks anything of you again, a commercial connection is the correct choice and there is no shame in it. Community ownership is a commitment of attention, not only of money.

How To Start Without Building An ISP

The common mistake is starting with wires. Trenching, pole agreements and transit contracts are the hardest and slowest parts of the problem, and groups that begin there usually run out of energy before anything works. Start at the other end.

Start with services. One machine, one domain, five people who agree to use it. Shared files, a calendar, a chat room, a video call that does not run through a corporate platform. It costs little, teaches the group how to maintain something together, and produces a benefit people can feel immediately.

Then link two buildings. A single wireless hop between two rooftops that already trust each other is the whole mesh problem in miniature: mounting, power, addressing, and who has the ladder. Get one hop through a full winter before planning twenty.

Then formalise, then grow. Incorporate, write bylaws, open a bank account and set a membership fee before expansion, not after. Growth applied to an informal group is what turns friends into a dispute over who owns the switch. Our other guides cover the pieces in more depth.

Where WOPR Fits

The WOPR Foundation is a 501(c)(3) nonprofit building community-owned mesh networking, self-hosted tools and direct social services. We supply pieces rather than running any single town's network, because the durable version of community internet is one a local group operates for itself.

In practice that is three things. We run and document self-hosted services, so a group does not have to work out from scratch how identity, mail and file sharing fit together. We support the mesh side, including the low-power radio layer that keeps working when everything else is down. And we build tools for the governance half, since a co-operative that cannot run a trustworthy vote will not survive its first serious disagreement.

We are also small, and it would be dishonest to imply otherwise. We are not going to trench fibre down your street. What we offer is working software, hardware that has been used in the field, and straight answers about what this actually asks of the people who take it on.

Frequently asked questions

What is community owned internet in simple terms?

It is internet infrastructure owned by the people who use it instead of a distant company. A co-operative, nonprofit, town or neighbourhood group holds the equipment, members decide together how it is run, and any money above operating cost goes back into the network. Think of a housing co-op rather than a landlord, applied to cables and servers.

What is the difference between municipal broadband and a community mesh network?

Municipal broadband is built and run by a city or public utility, usually as fibre, funded through bonds or utility revenue and staffed by employees. A community mesh is built by volunteers using wireless radios on rooftops, with no municipal involvement. Municipal networks are faster and better supported. Meshes are far cheaper to start and need nobody's permission.

Do I still need a regular ISP if my neighbourhood has a community network?

Usually yes, somewhere in the chain. Community networks own the local part of the path, but nearly all of them buy upstream capacity from a larger carrier to reach the wider internet. What changes is that the group buys it once and shares it, rather than every household holding a separate contract. Purely local traffic never touches it.

Is community owned internet legal in the United States?

Building a co-operative or nonprofit network is legal everywhere, though you need the usual business registration, insurance, and agreements for pole or rooftop access. Municipal broadband is different: several states have laws restricting or effectively banning cities from selling internet service. Check your state law early, because it determines which models are open to you.

How much does it cost to run a community owned network?

Costs split into capital and operating. Capital is the one-time hardware, mounting and setup. Operating is upstream transit, electricity, domains, insurance and a replacement fund for equipment that eventually fails. The mistake groups make is funding the first and forgetting the second. Savings against a commercial provider usually appear after a few years, not immediately.

How many people do you need to start a community owned internet project?

Fewer than most expect to begin, more than most expect to sustain. A shared server for files and chat can start with three or four committed people. A wireless network needs enough volunteers that losing one does not stop maintenance, which in practice means several people who can each handle roof work and basic troubleshooting.