#OMN is not trying to replace servers with #P2P, that’s never been the idea. The interesting architecture is hybrid, where each layer does what it is good at – #P2P isn’t the answer. It’s a tool. Browser-based P2P has hard limits, a browser isn’t a database, phones aren’t reliable servers, people go offline, storage is limited, connections are intermittent. If you build a system assuming every device is always available, it simply won’t work. The path we need is that #OMN isn’t trying to replace servers, it’s trying to change what servers do.
- Servers are good at availability, are online all the time, easy to reach. They’re ideal for caching, indexing, discovery, search and temporary message retention. They’re also cheap enough that communities can run them. Trying to eliminate them is mostly ideology rather than engineering.
- #P2P shines at resilience and autonomy, your phone is your primary editing environment, it stores your drafts, your identity and trust relationships. It can synchronise directly with nearby peers or trusted nodes whenever they’re available. That doesn’t mean every post has to live only on your phone forever.
Think “edge first”, not “server free” the #OMN architecture is closer to Git than to BitTorrent, many copies, many caches, many paths. No single point of failure, servers cache because they’re convenient. Peers synchronise because they produce the data, archives preserve because history matters, these are different jobs.
#ActivityPub already points this way, the Fediverse demonstrates the pattern when instances cache remote posts, users move between servers and archives exist independently. Nothing says there has to be exactly one canonical copy.
OMN simply pushes this idea a little further – Storage is a social problem as much as a technical one. The biggest mistake engineers make is assuming technology alone solves trust, it doesn’t, communities preserve history, archives preserve movements. Servers preserve availability, but people preserve the meaning. The goal isn’t to eliminate hosting, the goal is to avoid dependence on any single host, this is where the #4opens become architecture rather than philosophy.
Open Data means your data can move.
Open Standards mean different storage methods interoperate.
Open Source means anyone can build another client or server.
Open Process means communities decide how long things are retained, what is archived, and who they trust.
The path isn’t pure #P2P or pure client-server, it’s a federation of caches, archives, peers and communities, each doing what they’re best at. That’s a much more resilient model than pretending one technology can solve every problem.
The #OMN insight, that distinguishes the path from most “P2P” discussions is the primary object isn’t the file or the message – it’s the trust network. Content can be cached almost anywhere, but trust (identity, moderation, reputation, governance) is what needs to remain under community control. That shift – from moving data to maintaining trust – is where #OMN differs from most previous #P2P projects.
Resources for current #P2P
https://openhaven.net/prototype
https://www.localfirst.fm/landscape
