We're not doing the WoT thing. That's not our style. 🤙🏻
If you can get on a relay I use, I will see you.
Discussion
Wotathon ends in April, please be patient
My experience of WoT is that everyone's feed gets boring as fuck and they say "I love it here! Everyone is so nice here! ☺" and then they go back to X because their feed here is boring as fuck.
The mild.
Admonishing the mild.
To be milder.
I don't know which client here doesn't have WoT, maybe that's why they go back to x, so it's like but this is a bit like x but worse and then they choose the original, but if your choice is between mastodon and nostr client, you don't really know and it goes in that direction you want to block that instance because it has a low rating
Well, follow feeds are also a form of WoT, really. Same effect.
The WoT people are building here is very different from what I think of. It's some kind of social credit score determined by very important people.
WoT shouldn't be an algorithm on the graph, it IS the graph.
Einheitsbrei.
?Eggs and bacon? I am confused.
The state of the contents being uniform, plain, and indeterminate, like with porridge.
Are you saying that is how it is or how it should be? Still confused.
How WoT will cause it to trend. It's a way to trim off everything that isn't uniform.
Ah. Yup. WoT with algorithms requires that every assertion fit a preexisting category. You could do something with embeddings but that can only really interpolate between existing ideas. I just don't think we should have to wonder what we are missing.
We can and should filter who we talk to but we should leave filtering content to our brains.
You really don't know what they're doing if you don't follow them on Thursdays.
I'm not following closely.
Degrees of separation graphs and score graphs are two different things
This isn't true because you can "create" a database with your point of view or use a global one, or at least a more structured one.
Degrees of separation is always local and approximates an aspect of real-world trust directly. Score graphs are a much much rougher approximation that I think cannot reproduce the underlying signal. You almost need a global view to do it which means we are back to trusting an authority.
yes this is basically nip-85 you are trusting a provider to get data