french.toast'); DROP TABLE Messages;-- on Nostr: > Nostr is great at being a open protocol. But that also makes it highly suspectable ...
> Nostr is great at being a open protocol. But that also makes it highly suspectable to attacks, and spam. What I really want is decentralized identities like nostr but also the convenience of sites like reddit or lemmy.
>
> Lemmy’s main problem is that when one instance defederates from another instance all users are cut off from all users. If one user still wants to see content from OffensiveInstance A, but hes signed up in Instance B and B defederates from A he can’t.
just to play devils advocate a bit.. Do you think someone could come at the problem from the lemmy side and say 'fork this', starting with lemmy as a base so they get a lot of the ui stuff and whatnot but work under that framework on things like decentralized identities / making mod actions a per-user opt-in situtation? Not suggesting to do actually that, just curious what you think on chances of success if someone decided to try it that way
My guess is that there would still be a lot of issues (like how so many lemmy sites seem to have a lot of stability/network problems associated with mass federation) and it would be a big challenge due to existing assumptions in lemmy codebase. Plus - unless whatever fork with the changes actually managed to gain enough momentum to overtake the original - most would just ignore it due to the network effort.
>
> Lemmy’s main problem is that when one instance defederates from another instance all users are cut off from all users. If one user still wants to see content from OffensiveInstance A, but hes signed up in Instance B and B defederates from A he can’t.
just to play devils advocate a bit.. Do you think someone could come at the problem from the lemmy side and say 'fork this', starting with lemmy as a base so they get a lot of the ui stuff and whatnot but work under that framework on things like decentralized identities / making mod actions a per-user opt-in situtation? Not suggesting to do actually that, just curious what you think on chances of success if someone decided to try it that way
My guess is that there would still be a lot of issues (like how so many lemmy sites seem to have a lot of stability/network problems associated with mass federation) and it would be a big challenge due to existing assumptions in lemmy codebase. Plus - unless whatever fork with the changes actually managed to gain enough momentum to overtake the original - most would just ignore it due to the network effort.