The anecdote I love to give is that I didn't know that Brin went to my high school until after I'd graduated. It's a high-performing public school due to its proximity to several research institutions, but it was never exactly loaded, and certainly could have benefited from outside investment (say, to replace the 20ish "temporary" trailers with a new wing). Even just having him show up to give a talk to students would have been amazing. Not a peep from this man, though, let alone the pocket change to help out his alma mater.
This is the flip side of the "self-made man" narrative.
It allows one to disavow any sense of social reciprocity after becoming obscenely rich.
I was curious, so I looked through his Wikipedia page -- it says he donated $1m to the Hebrew Immigrant Aid Society in 2009 (which helped his family move to USA when he was a child). Even the NYT article notes that "The gift is small, given Mr. Brin’s estimated $16 billion in personal wealth" :D
(this is like you making $1m annually and donating $62.50)
Network routes consist of a network (a range of IPs) and a next hop to send traffic for that range to.
These can overlap. Sometimes that’s desirable, sometimes it is not. When routers have two routes that are exactly the same they often load balance (in some fairly dumb, stateless fashion) between possible next hops, when one of the routes is more specific, it wins.
Routes get injected by routers saying “I am responsible for this range” and setting themselves as the next hop, others routers that connect to them receive this advertisement and propagate it to their own router peers further downstream.
An example would be advertising 192.168.0.0/23, which is the range of 192.168.0.0-192.168.1.255.
Let’s say that’s your inference backend in some rows in a data center.
Then, through some misconfiguration, some other router starts announcing 192.168.1.0/24 (192.168.1.0-192.168.1.255). This is more specific, that traffic gets sent there, and half of the original inference pod is now unreachable.