The Internet has become one of the most important pieces of modern infrastructure—but it’s also built on a surprising amount of centralized trust.
Every time you connect online, you’re relying on Internet service providers to assign addresses, DNS servers to translate domain names, certificate authorities to verify identities, and registrars to determine who owns a domain. Most users never think about these systems because they work remarkably well. But they also introduce points of control, surveillance, censorship, and failure.
What if networks didn’t need any of that?
That’s the idea behind two fascinating open-source projects: Reticulum and FIPS. Both aim to create decentralized networks where devices can discover each other, exchange data securely, and build resilient communication systems without depending on traditional Internet infrastructure.
Although they share a common vision, they take very different paths to get there.
Rethinking the Internet’s Foundations
Both projects start with a simple principle: a device shouldn’t need permission from a central authority to join a network.
Instead of requesting an IP address from a provider or relying on a trusted third party to establish identity, each device generates its own cryptographic key pair. That cryptographic identity becomes the device’s permanent address on the network.
Because identities are derived from cryptographic keys rather than assigned by an organization, there’s no central registry to manage, no certificate authority to trust, and no administrator required to approve new participants.
This doesn’t just simplify deployment—it also makes the network far more resilient to censorship and infrastructure outages.
The Network Doesn’t Care How You Connect
One of the most interesting design choices shared by both Reticulum and FIPS is that neither one is tied to a specific communication technology.
Whether two devices communicate over:
- Wi-Fi
- Ethernet
- Bluetooth
- LoRa radio
- Serial cables
- Satellite links
- VPN tunnels
- Or even the public Internet
…the protocols don’t really care. As long as data can move from one device to another, it can become part of the mesh.
This makes the projects useful for a wide variety of scenarios, from emergency communications and disaster recovery to remote sensor deployments, community mesh networks, and off-grid experimentation.
Security Comes First
Unlike traditional networks, where encryption is often added by individual applications, both projects assume that the underlying transport is completely untrusted.
Whether traffic is passing through a public Wi-Fi network, bouncing across multiple radio links, or traversing someone else’s infrastructure, communication remains cryptographically protected.
Each node can verify the identity of the other without relying on centralized certificate authorities, and messages remain encrypted end-to-end.
In other words, the physical network simply becomes a transport mechanism—the security lives within the protocol itself.
Where the Projects Begin to Differ
Although their goals overlap, Reticulum and FIPS are optimized for very different environments.
Reticulum: Built for the Impossible
Reticulum was designed with one question in mind:
How little bandwidth can a network realistically operate on?
The answer is astonishingly low.
Reticulum is capable of functioning over extremely constrained radio links where bandwidth may only be a few bits per second. To make this possible, every part of the protocol is engineered to minimize overhead and avoid unnecessary transmissions.
Rather than building on top of existing Internet protocols, Reticulum introduces its own networking model specifically designed for these conditions.
The result is a protocol that’s exceptionally well suited for:
- Long-range radio communication
- Emergency and disaster response
- Remote environmental sensors
- Off-grid communities
- Experimental networking projects
Its philosophy is simple: if communication is possible at all, Reticulum tries to make use of it.
FIPS: Bringing the Existing Internet Along
FIPS takes almost the opposite approach.
Instead of replacing familiar Internet protocols, it allows the decentralized mesh to appear as a standard IPv6 network.
Why does that matter?
Because existing software already understands IPv6.
Rather than rewriting applications to support an entirely new protocol, many existing tools can continue operating with little or no modification while the mesh handles routing behind the scenes.
This dramatically lowers the barrier to adoption, making FIPS attractive for organizations and developers who want decentralized networking without abandoning decades of existing software.
Two Different Philosophies
In many ways, the projects represent two schools of thought.
Reticulum asks:
“What would networking look like if we started from scratch and optimized for resilience above everything else?”
FIPS asks:
“How can we build decentralized networking while remaining compatible with the software people already use?”
Neither approach is inherently better.
Instead, each reflects a different set of priorities.
If your network must survive with minimal infrastructure and almost no bandwidth, Reticulum is difficult to beat.
If compatibility with today’s networking ecosystem is more important, FIPS provides a much smoother transition.
Why This Matters
Discussions about decentralized technology often focus on cryptocurrencies or blockchain projects, but networking itself remains one of the Internet’s most centralized components.
Projects like Reticulum and FIPS remind us that communication doesn’t have to depend on large providers, centralized identity systems, or globally coordinated infrastructure.
As concerns around censorship, surveillance, Internet shutdowns, and infrastructure resilience continue to grow, alternative networking models are becoming increasingly relevant—not because they’re trying to replace the Internet overnight, but because they offer another option when traditional assumptions no longer hold.
Whether either protocol becomes widely adopted remains to be seen. But both demonstrate an important idea: secure, decentralized communication is no longer just an academic concept. It’s something you can experiment with today.