siwek (@siwek@social.tooinconsistent.com)

@_elena@mastodon.social It seems they are calculating "engagement" which I'm assuming is likes + reposts + comments and then show it as number of comments.

we can just do things - underreacted

7 days later and we have some results from the experiment. When we demote popular posts we see: - 8.26% fewer "show less like this" (3340 -> 3064) - 0.24% more posts in For You were liked (242438 -> 243024) - 2.43% more feed loads (438867 -> 449537) Per user and per request metrics:

Value misalignment in X’s feed algorithm is a reflection of value tensions in engagement

Social media feed algorithms rank content that is purported to be preferred by users, but the engagement behaviors that drive these algorithms are (at best) indirect proxies for users’ explicitly self-stated values. Are the resulting feeds value aligned, and if not, why? We investigate this question by annotating the basic human values expressed in participants’ X (Twitter) feeds (N = 715 US users), analyzing the relationship between the posts’ value expressions and the posts’ amplification in the ranked “For You” Page feed, and then comparing the amplified values to users’ own values. We observe that the inventory of posts from followed accounts reflects users’ self-stated values—but that there is an overall negative correlation (misalignment) between users’ explicit values and the value expressions the algorithm is more likely to amplify. We turn to engagement behavior to understand this misalignment and observe that users’ engagement behaviors can be misaligned with their stated values—likely causing the algorithm to learn and reflect these misaligned values. We also detect partisan differences consistent with this theory: While the algorithm amplifies values negatively correlated with both Democrats’ and Republicans’ self-stated values, they are more misaligned for Democrats. And in fact replying, a heavily weighted form of engagement, is associated with values that are less aligned for both Democrats’ and Republicans’ self-stated values, and is even more misaligned for Democrats. Taken together, these findings offer a glimpse into the tensions between the values that people hold and those that provoke reactions, and how these value tensions can produce misaligned outcomes.

Value misalignment in X’s feed algorithm is a reflection of value tensions in engagement

Social media feed algorithms rank content that is purported to be preferred by users, but the engagement behaviors that drive these algorithms are (at best) indirect proxies for users’ explicitly self-stated values. Are the resulting feeds value aligned, and if not, why? We investigate this question by annotating the basic human values expressed in participants’ X (Twitter) feeds (N = 715 US users), analyzing the relationship between the posts’ value expressions and the posts’ amplification in the ranked “For You” Page feed, and then comparing the amplified values to users’ own values. We observe that the inventory of posts from followed accounts reflects users’ self-stated values—but that there is an overall negative correlation (misalignment) between users’ explicit values and the value expressions the algorithm is more likely to amplify. We turn to engagement behavior to understand this misalignment and observe that users’ engagement behaviors can be misaligned with their stated values—likely causing the algorithm to learn and reflect these misaligned values. We also detect partisan differences consistent with this theory: While the algorithm amplifies values negatively correlated with both Democrats’ and Republicans’ self-stated values, they are more misaligned for Democrats. And in fact replying, a heavily weighted form of engagement, is associated with values that are less aligned for both Democrats’ and Republicans’ self-stated values, and is even more misaligned for Democrats. Taken together, these findings offer a glimpse into the tensions between the values that people hold and those that provoke reactions, and how these value tensions can produce misaligned outcomes.

sample of 50 replies, mentions, and QPs

Follows & Replies by @mackuba.eu

Feed that shows all posts from people you follow and all their replies to anyone – kind of like in the original Following feed (though without reskeets). Excludes replies to you, to the same user, and replies made in your own threads. (See also the "Only Replies" and "Only Posts" versions.)

Elena Rossini 🌈 (@_elena@mastodon.social)

Attached: 1 image 🚨 Important update about my latest #WSocial article 👀 A user on #ATproto cracked the code and came up with a reasonable explanation for the made up number of comments on W Social's homepage. The number next to the speech bubble (comment icon) are an "engagement metric": the sum of boosts and likes. It all makes sense now! https://blog.elenarossini.com/w-social-fictional-metrics-and-the-beauty-of-open-data/ (scroll down to the end of the article for the update)

Elena Rossini 🌈 (@_elena@mastodon.social)

Attached: 1 image 📝 #WSocial state of the network - Friday July 10th 📣 for those who are curious about it, considering the enthusiastic endorsement by prestigious European institutions #OpenData allows me to crunch numbers. On day 23 since its beta launch in Brussels, W Social has 11291 accounts: - 469 new users since yesterday - 899 users have published one or more posts - 72 new users posting since yesterday - 92% have never* published a single post *may be related to #WIdentity and its 13-step process

spacecowboy (@spacecowboy17.bsky.social)

PSA: you can find out which of your likes caused For You to show any given post at foryou.club/explain Just enter the post URL and hit "Explain":