[nullc]

> You send a payment to the next hop in the route, then they send a payment to the following hop, and so on.

It's not technically a payment at that point, since the payments is atomic end to end. But yes, you send a message your peer, which sends it to another peer, which sends it to another peer...

like any other P2P network.

[shilch]

I don't think that "Peer-to-Peer" in the whitepaper's title ("Bitcoin: A Peer-to-Peer Electronic Cash System") refers to the network structure being p2p - and that's probably also not what most people mean when they talk about bitcoin being peer-to-peer. The very first sentence makes it pretty clear that peer-to-peer means person-to-person without any intermediary:

"A purely peer-to-peer version of electronic cash would allow online payments to be sent directly from one party to another without going through a financial institution." [1]

Furthermore, the previous title was "Electronic Cash Without a Trusted Third Party" [2]. So reading "Peer-to-Peer" as "Person-to-Person" would mean that the title hasn't changed in meaning, it just became a bit more catchy.

Also, when analysing the incentives of participants in the bitcoin network, it turns out that network nodes do not actually form the ideal-typical p2p mesh network (where all nodes are equally distributed and connect to a few other nodes) but a more densely connected network where connectivity to mining nodes is strongly incentivised. This topic has been researched and discussed by Dr. Wright (See [3] for more information).

[1] https://bitco.in/bitcoin.pdf

[2] https://nakamotostudies.org/literature/ecash/

[3] https://nchain.com/en/blog/bitcoin-network-topology-small-wo...

[nullc]

Yes, and lightning does the same. The other bitcoin nodes that route your messages are not a trusted third party, no more than the bitcoin nodes that relay your transaction when you transmit it any time you use Bitcoin.

Wrights comments on topology are technobabble and largely meaningless, so it's difficult to say something about them... however, to the extent that we can assign any meaning at all to them don't you notice that saying your transactions need to go through particular nodes sounds an awful lot like the property you're using to argue that lightning is not peer to peer?

[phillipsjk]

Unlike the base layer, the LN uses source routing: which was abandoned years ago. Requiring the endpoints to know the structure of the network graph leads to intractable routing problems at scale.

[nullc]

Many source routed networks work fine, including Tor. All MPLS usage is source routed. Source routing isn't used for _IP_ (generally) but for reasons which have nothing to do with 'scale'.

[etherael]

Not a problem if you don't intend to have nodes sufficient to actually be considered "at scale" and your plan from the very beginning is a centralised hub and spoke network.