some photos here https://medium.com/@ksenia_1915/прекрасные-интернеты-будущего-интервью-с-создателем-протокола-cjdns-b081742873e8

I has some ideas on this point, but I didn't find a solution about how we can spread supernodes globally on the globe.

But I find a way how to implement bandwitch market in some basic nature.

Sender (offline) can simply attach transaction with it's message like  a "bank check". The first one in network who submit message (online access) execute transaction

cjdelisle [matrix]: Problem with vlang AFAICT is it's not memory safe

cjdelisle [matrix]: If I'm going to use a non-memory-safe language, it might as well be C

cjdelisle [matrix]: anyway, nap time

Plus some smart-contract, which can check hash for submitted message and hash of message, that we transfer inside of transaction. This how possibly validate message and give sender a guarantee of message delivery

IMO any blockchain-like payment system is incompatible with small and fragmentating networks...

hehe, did the tree make any noises when it's fall if nobody in the woods?

I mean sidechains possibly can work with fragmentating networks. Other solutions from layer two can work with it as well I suppose.

If network starts fragmentation it can create just a new branch inside of it.
Mmm... imagine that tommorow russian goverment will cut of runet from internet, what will happens to crypto?
Russian miners network reconfigure itself, and payments will working, but now it will be unnatural and unwanted fork of blockchain if we are talking about bitcoin for example

Sharding, sidechains and other solution of second layer can do this process more stable and effective in small networks and thin networks as well

Okay, so my point of a view - we actually can build a bandwitch market on a mesh network on Ethereum blockchain, I know how to build it

But there are coming a few problems like "how to motivate edge nodes / supernodes" or "how to make independent homeservers deployment accesable to a dummy user" or "how to make a trully distributed federation" or even just "who exactly is gonna pay for bandwidth"

In a some way I actually know how to make user phone communicate with router and send direct pay for router for accessing network, but it seems impossible to propagate this idea just like that.

I think firtsly local organisers will pay for a bandwidth or just a provide free internet access point for everyone in exchange for subscription for an organiser work.

Like you came to a coffe-shop, join network, subscribe to a local chat and have a free network access. Coffie-shop get a new community member and a new customer, so it is motivated to have a local mesh node always online. Users in this case get free network access without neccesarity to pay

Other way we could just simply implement a bandwidth chain from offline location to online location with microtransactions for every routing node in a chain

BOOM! https://vlang.io/play

openlet [matrix]: Test

passed

user1234 [matrix]: failed

That is exactly what I am thinking of last years

I have an experience in the past in campus network building ( homenet's ), friends-based FTN networks and so on

question is - it never last longer that it had a reason for participants

for campus networks back then it was a internet access sharing

for ftn - community which existed only there, as well as an access model ( quite hierarchical )