I also hope you don't intend to communicate anything that requires security or privacy because, assuming you mean HAM bands in HF, then you can't encrypt anything.
The problem with using RFC1149 is rounding up and training enough seabirds—pigeons won't fly that far over water. Albatrosses would be ideal, but there might not be enough of them available.
Hmm. Can we instead attach mesh repeater units to the albatrosses, and hope that enough of them take up optimum spacing for long-distance transmission? Or perhaps it would be better to just string a line of buoys across the Bering Strait once Russia stops being cantakerous. Then all we have to do is find a way to connect up Australia and a few assorted islands here and there.
From what I understand it's the final form of decentralization. The idea is that a bunch of user/community owned computers communicate with each other and act like an alternative community based internet provider. Big cities already implement some form of this like NYC.
How would you propose connecting to a mesh network without the use of an ISP? Are you thinking wireless (wifi)? For something like that to work you would need an AP/repeater every 150 feet or so. How would you cross oceans without bouncing wireless signals off the
ionosphere (creating latency issues) or using a cable (requiring some entity to maintain it - $$$)?
If you break it down into the requirements to do such a thing you can start to understand why your question is a bit ridiculous.
I remember reading about a "guerilla wifi" mesh network in NYC, and I did a bit of research on connecting to that several years ago. It turned out I was too far away from Manhattan to be in range. But also from what I read a series of small-scale peer-to-peer connections don't give you the low-latency or throughput of a good backbone.
Actually I'd like to add a note about how much I appreciate infrastructure. It would be great if we could all equally own and control the Internet. But when you get down to it, societies pooling their efforts can do things that small, independent groups can't, such as building tier 1 network backbones.
Looking at it another way, if you did have a global mesh network it would be made up of electronics that take tremendous systems of supply chains and factories to build and distribute. That's sort of the same idea: large-scale infrastructure that small groups can't pull off.
If I had my way I would keep the large-scale networks, but change the governance model to shift the primary organizational motivation from profit to human wellbeing.