jb55's avatar
jb55 _@jb55.com 3 months ago
block the ones controlled by a single entity yes, so its a more accurate sample of real node runners. it appears to be much less (2-3%) instead of 20%. so some scammer is inflating the numbers to convince people that users are running knots. honestly could be a state level actor at this point, who knows.

Replies (6)

Could be the individual I overheard scheming to inflate the node count numbers not too long ago right before the knots count started to spike. 🤔 Node count is a largely irrelevant vanity metric. Best to ignore.
FWIW it’s not quite so simple. It’s really gonna bias towards small ISPs. Comcast may have 300 nodes but you’re gonna bias towards the nodes on random networks across the world or small hosting providers with one node. I do agree that it’s a better sample (mostly cause it’ll also bias away from OVH/GCloud/Hetzner), but I’m not sure it’s great either.