The algorithm is killing Twitter and it's driving me insane
hikari.noyu.me
hikari.noyu.me
The algorithm always tried to introduce you to people you might find interesting. This would obviously work by looking at who you follow and coming up with similar matches.
But what happens when the site is losing users day after day? The algo can’t feed you your regular tweets, half of them are no longer posting. Worse, half the people it would have recommended have gone as well. So, it casts its net far wider than it used to. This picks up a whole bunch of tweets that, for whatever reason, you have no interest in reading.
And yes, in the year after the takeover, I found my feed was just, increasingly, boring. Then some technical screw up made it not work on my phone browser and… I just couldn’t be bothered to download the app and work around it.
Twitter’s product was always the people, and they’re not there any more.
My experience on the website hasn't changed, but I deliberately only follow accounts that do not engage in news or politics. My account is completely detached from current events besides my local municipality's accounts.
After Elon purchased Twitter, I suspect things bifurcated after a large contingent of left-of-center people rage-quit. This created two asymmetries: one on Twitter, and one on Mastodon. Now there are two quasi-echo-chambers that to go together oil and water. Any time someone tries to cross over, they experience what appears like a uniform attack against them, but in reality it's because people took sides and sorted themselves into separate networks.
The woodworking communities on Twitter and Mastodon are both excellent and different.
I’d argue that a persons preferences on what they want to not be screamed at about on the internet is actually a great thing to take sides in. After all, it’s why we have more than one social media platform.
That post is about as articulate as I’ve seen that explains why some people don’t want visibility, or who go “death con 3” (after Kanye West) when they see a reply they don’t like. There’s a real contradiction between the author’s desires and having a big pool to swim in, algorithmic feed or not.
Hi, I am actually the author of the post, but I tend to keep a low-ish profile here. I think you may have misread me a bit, or I didn't make myself clear. I want to explain why as I think it may be insightful.
The thing that's bothered me for a long time now about the fediverse is that it has this culture of inter-instance suspicion. Fediverse users seem to expect instances to correspond to roughly what I call “subcultures” in the post, and then form the moderation/federation policy based on that, cutting you off from other subcultures that aren't aligned with “us”. Even if this doesn't happen at a technical level, people seem to act that way at the social level. A lot of weight seems to be placed on what instance you're on.
To me I can't help but find this deeply toxic. I actually have a single-user fediverse instance, and I suppose in some sense that means I have the “smallest and queerest” Mastodon server, but it's really very different. I did this because I refuse to belong to just one subculture. I think every person is part of many different little subcultures at the same time in their different spheres of life, and I really didn't want some instance admin deciding for me which cultures I am and am not allowed to be part of. I also didn't want people to immediately dismiss me as being from the “wrong” culture by my handle. I guess I'm one of those people who think friendships are somewhat sacred and social pressure to make all your friends be culturally aligned is not good.
As much as I said in the post that the lack of legibility in the audiences on Twitter with the current algorithm is bad, I think excessive legibility can also be bad. Humans are too tribal. The fediverse puts one particular tribal marker front-and-centre and it seems to break too many people's brains at some level.
Having to keep tabs on both Twitter and the fediverse already takes up a lot of my attention, so I haven't really felt able to maintain a serious presence on other places, but it seems like Bluesky might be closer to the Twitter philosophy here, so I'm more optimistic about that site if I had to pick a single succcessor to Twitter.
> That post is about as articulate as I’ve seen that explains why some people don’t want visibility, or who go “death con 3” (after Kanye West) when they see a reply they don’t like. There’s a real contradiction between the author’s desires and having a big pool to swim in, algorithmic feed or not.
I want to push back on this too. I don't think that no visibility between these groups is good, and I also don't think hyper-visibility is. I think what I'm arguing is that there's a happy medium that's being lost, or that the particular compromise Twitter had made Twitter specifically work.
[Added in an edit:] There's definitely some small and super insular subcultures that lose out from this constant bubble-bursting effect, but it's not just those that do. There's also for instance a pretty huge subculture I'm part of on Twitter that's defined by people being, more than anything else, open-minded and willing to assume good faith. That subculture doesn't hate contact with others, it really appreciates diversity, but for its own survival it needs to keep at least some distance from the people on Twitter who are so terminally tribal that they will attack them on sight.
I have to figure out how to tell which of my followers is an actual well intentioned human, and purge the rest
I miss Reddit, having nuked my account, and vowing never to go back.
The only tenable future is going back to Blogs and RSS, where there's no algorithm, for the most part.
Just curious, why?
I'd dump all social networks if I could, but my family is at Facebook.
In my case there were too many negative opinions affecting my mood, and I was spending too much time on the site
Now, no matter what, I get inundated with far right, and sometimes viciously racist drivel. Elon himself seems inescapable - blocking him stops the (completely unwarranted) push notifications of "wow i can't believe this" grandma-emailing-me-something-she-found-in-a-FW:FW:FW:FW:-chain style of tweets he will constantly post, and to a large extent keeps him out of my feed - but then there's a blue-check army of sycophants that will not just quote tweet but literally screen shot and repost his tweets that will start showing up in my feed too. I've tried and failed to stop any of this. To be fair, I didn't try super duper hard (because I shouldn't have to), and I'm sure there's some way to tailor it that I don't care to figure out. I decided I simply don't need to be there, and if that stuff floats your boat, that's fine by me - I just don't want to see it.
Really though, the biggest problem to me when engaging on platforms, especially if I accidentally "blow up" like the author describes - people have this pathological need to be angry at something or someone. Like, I've been in these baffling exchanges with people where I basically throw my hands up and say "Look, you don't have to be here engaging with me, you don't have to read anything I write, you can just go on about your day" but it becomes clear they are doing it because they enjoy that outrage and harassment. And that problem is definitely not unique to twitter/x, although some platforms deal with it better than others.
Yes. I think this is corrosive to social media sites. It's also something that kinda defines them of course. If they never got anyone animated for any reason, they'd be boring, but there's some threshold where it starts making things a bad time.
That is, they want to be able to post things that show vulnerability without having the posts be visible to people who will attack people that show that vulnerability. They want less visibility.
For once somebody like this is expressing themselves in a blog and actually expressing themselves. It still seems to me to be a naive position; there is the fight over blocking in Twitter right now where people are claiming they need blocks so that their enemies can’t read their posts but what stops their enemies from making another account? (I guess the block still stops replies though)
If people are that sensitive however they probably should be in a closed space of some kind.
For example, reading the conversation that follows from a single tweet. There's pages of culture war bullshit, spam, AI, paid blue checks that all gets ranked more highly in the replies than what quality discussion there might be.
Another example, let's say you post a tweet. The tweet is shown to others who aren't following the strict chronological timeline of their own followers. The "algorithm" (whatever that is) shows it to people who hate your stance, because that's what drives engagement/revenue, and now your own replies are flooded with shit.
I've heard the complaints that twitter was ruined, but never understood how. Author points to a loss of personal community, so now I think that the complaints are a result of Musk's deliberate bursting of filter bubbles, and are from people who preferred to keep their bubble.