Lightnews — Scholar-powered news

spacecowboy @spacecowboy17.bsky.social · 7h

There are a few parameters to tweak - how much to increase/decrease the scores. This could be tested out.

spacecowboy @spacecowboy17.bsky.social · 7h

The starting conditions are: you have some nominal amount of "like trust" for every user and your "dislike trust" is 0 for every user. Which means that for a new user the reply section is sorted by like counts.

1 1

spacecowboy @spacecowboy17.bsky.social · 8h

Posted the idea: bsky.app/profile/did:...

spacecowboy @spacecowboy17.bsky.social · 8h

Here is an idea how to make the discourse healthier: change the incentive structure of the reply section.

Right now the replies are sorted by the number of likes. This is extremely gameable. 🧵

spacecowboy @spacecowboy17.bsky.social · 8h

3. as a secondary effect, posters/likers would know that their actions could have negative consequences for the amount of attention they will be getting in the future

1 2

spacecowboy @spacecowboy17.bsky.social · 8h

The result is that:
1. each user gets agency over how their future attention will be allocated in the reply section
2. we don’t have to resort to the blunt tool of blocking users

1 4

spacecowboy @spacecowboy17.bsky.social · 8h

When calculating the score of a reply we subtract from it how much you trust the dislikes of people who disliked this reply.

This may make this reply have a negative total score. In that case it can be hidden or grayed out.

1 2

spacecowboy @spacecowboy17.bsky.social · 8h

When you dislike a reply - you increase your "dislike trust score" for people who also disliked that post.

When you like a reply - you decrease how much you trust the dislikes of people who disliked this reply.

2 2

spacecowboy @spacecowboy17.bsky.social · 8h

Then when sorting the replies we sum the "like trust scores" of people who liked that reply.

There is another set of "trust scores" - how much you trust the dislikes of other people.

1 2

spacecowboy @spacecowboy17.bsky.social · 8h

Note: we treat the author of the reply as a liker.

We also need a dislike button (kind of like "show less like this" in feeds) - through which you can decrease your "trust score" for people who liked the content you didn’t like.

1 3

spacecowboy @spacecowboy17.bsky.social · 8h

We can fix that by creating a feedback loop that rewards good behavior and penalizes bad behavior. Who is the judge of deciding what is good vs bad, you ask? You are the judge.

When you like a reply - you increase your "trust score" toward people who liked that reply.

2 10

spacecowboy @spacecowboy17.bsky.social · 8h

Here is an idea how to make the discourse healthier: change the incentive structure of the reply section.

Right now the replies are sorted by the number of likes. This is extremely gameable. 🧵

3 3 23

spacecowboy @spacecowboy17.bsky.social · 8h

The paper raises a lot of good questions.

@why.bsky.team is working on healthy discourse project and I think the paper could be useful.

I have an idea on how to make the discourse healthier through a better incentive structure in the reply section. Will share it soon.

1 1

spacecowboy @spacecowboy17.bsky.social · 1d

Great to hear! See: bsky.app/profile/spac...

spacecowboy @spacecowboy17.bsky.social · 4d

Before creating For You I actually built a link-focused personalized feed Links For You.

It uses the same algorithm but:
- it only looks at posts with links
- it groups all posts with the same link to avoid showing different posts that have the same link

Feedback welcome!

1 1

spacecowboy @spacecowboy17.bsky.social · 1d

Which goes to show that people have different definitions of toxicity.

It's worth discussing what we consider toxic. In my view toxic behavior is meant to make someone feel unwelcome.

1 1

spacecowboy @spacecowboy17.bsky.social · 1d

Try this For You - it does not try to optimize for engagement unlike many personalized feeds. It connects you to people who liked the same posts as you and shows you what else they've liked.

2 1

spacecowboy @spacecowboy17.bsky.social · 1d

Thanks! I didn't know such emoji existed. It does convey the idea of the feed. Definitely an improvement over the existing icon.

In my mind ✨ = "magic" = AI. For You is intentionally simple and I'd like to capture that somehow. I wonder what @designleviathan.bsky.social will come up with.

1

spacecowboy @spacecowboy17.bsky.social · 2d

Reminds me of this feed: bsky.app/profile/xove...

1 2

spacecowboy @spacecowboy17.bsky.social · 2d

I was thinking of simply taking a photo of a crochet heart which would have some softness and human touch to it.

1 2

spacecowboy @spacecowboy17.bsky.social · 2d

That would be great! The current icon tries to convey how your likes lead you to more content you will like. Though it looks too hard to read.

Maybe a single larger heart with some pattern that shows connection?

2 1

spacecowboy @spacecowboy17.bsky.social · 2d

Would anyone be interested creating an avatar for the For You feed in exchange for a credit in the feed description?

I have no idea if this is a fair offer.

4 2 20

spacecowboy @spacecowboy17.bsky.social · 2d

Here it is:

3

spacecowboy @spacecowboy17.bsky.social · 2d

Try using "show less like this" on posts you don't want to see in For You and it will less connected to people who liked those posts.

1 1

spacecowboy @spacecowboy17.bsky.social · 2d

Try For You - a personalized feed based on your likes.

spacecowboy @spacecowboy17.bsky.social · 2d

Try For You - a personalized feed based on your likes.

To curate the feed, like posts that you want to see more of and use "show less like this" on posts you don't want to see.

1

spacecowboy @spacecowboy17.bsky.social · 2d

Try For You - it shows art that people who liked the same posts as you liked.

1