Post #2564663
2024-12-19 22:18 UTC
@miblo@mas.to @rauschma@fosstodon.org @geertaarts@mastodon.green yep nice
I’m thinking … if we do the hashtags — the scores will be per message (sometimes). And these will sometimes be lower than the score for the account holder 1:1 relationship.
But, we’re doing hashtags to sink statements as well as surface statements — so, fine.
I’m heading for terminology now
- statement score — the final metric
- trust factor, cofactor — used for algorithm
- vouch value — manual overrides
- filter slider — your screen setting
Replies (1)
-
@miblo@mas.to 2024-12-20 02:31
@whophd@ioc.exchange @rauschma@fosstodon.org @geertaarts@mastodon.green Cheers, they will, yeah. Gotta say, by the way, I think the filter slider is a stroke of genius. I mean, the scoring idea from https://mas.to/@miblo/113581191464258423 is in the context of a system where the messages, I imagine, are laid out in 2D, by their proximity to each other and us. But in the 1D of Mastodon, there's no space for proximity, a message can only be or not be shown. So how on Earth do you pick what to show, given a score of 60%? Our friend the filter slider.