Sounds like a load of crap to me, lets have some maths to support the claims please.
The fact that an entire mining pool was on an orphan blockchain for more than 300 blocks isn't enough to convince you?
How many nodes represent one end of the chain to the other?
I'm not really sure what you're asking here. "Maximum end-to-end propagation time" is the maximum amount of time it would take for a message from any node to reach every other node. Each node is an "end" in that each node is both a source and destination for messages. For a given node, the "opposite" node is the node out of all nodes where sending a message to would take the longest.
Ever heard of the 6 degrees of separation theory?
Can you articulate why that theory is relevant here? In that theory, each person can have an unlimited number of connections. In Bitcoin distributed networks, each node has a limited number of connections (only eight by default). So again, can you articulate why that theory is relevant here?
How do you compensate for a node that has bad lag or off-line for half an hour? Well above any useful block rate.
The network is connected enough so that there is multiple ways to route messages between each pair of nodes. If a node dies on a route between two nodes, messages are redirected across other routes that don't involve the dead node.