Just came back to this paper. The core spanning tree that they describe sounds similar to the transport overlay network in this paper.

So I don't think they are too far off.

I would argue that it is very unrealistic to have n-1 used and balanced channels in a network with n nodes.

In reality everyone wants to increase their own connectivity. Even the most ""conservative"" people only want channels to peers with at least 2 other good channels. Because 3 channels are an increase in routes while 2 channels are just a straight line. But even looking past ""conservative"" users: there are so many people that try to earn money in Lighning and open/close their 20+ channels all the time to maximise gain.

The idea makes sense but I can't see that this is the world the Lightning network is moving towards.

Welp. That was an accident.

Idk about how I should feel about publishing "papers" on Github? Feels like it's blurring the line between academic work and oldschool in-depth Reddit posts.

Interesting paper! Where did you find it?