Very good ideas, but I don’t agree at all with your estimations.
For example it is terribly difficult to self host email, and very few people actually do it. Contrary to your estimation. It’s not because of the server software, but because of the fight against spam etc. that costs so much.
You are focusing on only one “top” and so you can’t see the reality. You are scoring centralization. Not decentralization.
Better if you look at the share of hosters at the “lower end”, the ones that actually do self hosting, like:
% of servers with up to 10 users (counting only natural persons).
% of users on servers with up to 10 users (counting only natural persons).
SufferingSteve@feddit.nu 2 days ago
I like the idea, but I don’t really feel this is scoring decentralization at all.
Yeah,looking through this again, it really scores centralization, and also focuses way to much on ease of setup. Which honestly has nothing to do with it. If it’s super hard to setup, but every participant is hosting their own node. And producing the same amount of content. That would be max decentralization.
I would also argue that a requirement for decentralization is that the service keeps working even if a large portion of all nodes goes down, the remaining nodes are still operational and keep delivering the value.