Twitter has a setting where you can specify words that if they appear in posts, they'll not show up on your feed. This helps me.

However, several other factors also help me:

- I'm very aggressive with telling Twitter and any other platform when I don't like something by using the "Not Interested" button or equivalent. This seems to work well with Twitter.

- Twitter loads slowly on my system, so I use a Firefox add on that hides the whole sidebar or whatever they call the trending topics. So I don't see them anymore, even though they weren't really bothering me in the first place.

- I honestly barely use Twitter except to see the latest on specific accounts. I'll usually click on links directly to access them and not depend on a feed. Stuff that is super popular on Twitter tends to be screenshotted and posted to other platforms anyway.

- Overall my even minimal use of Twitter is dwindling; most of the accounts that have content I like on Twitter are also on BlueSky.

It sounds like we’re converging on the same Twitter workflow over time, which is to just never use it at all, haha.

Of all the social media platforms, Twitter used to be the one that was least likely to make me angry or otherwise produce an unpleasant reaction.

I suspect that generally coincided with the ability to use it briefly and then walk away for a day or more. Which ... isn't good for ad impressions, and I suspect the main driver for engagement hooking.

A lot of people don't have the ability to reject this, it seems, and it obviously has a real mental health toll. The second I find myself feeling "bad" when using an app I just close it down for weeks at a time. No place has that been more true or for longer than the current iteration of Twitter. Fundamentally ad-driven social media works best financially when they have you there mashing buttons and clicking posts.

> It sounds like we’re converging on the same Twitter workflow over time, which is to just never use it at all, haha.

True, it boggles the mind the extents to which some people twist and turn to salvage the usability of a system that's compromised and managed so that it actively works against your best interests.

It's like enumerating a list of pro tips on what sprinkles to pour over shit sandwiches to make them less revolting. No, sprinkles are not a solution. The solution is to not eat shit sandwiches.

I have an extension that removes twitter blue posts and other advertisements.

I've tried to use Twitter on multiple occasions over the years. The thing that really annoyed me the last time I gave it a go (and once again ultimately decided there was nothing of value to be found there) was that it would show me posts from pages that I don't follow, but that people I follow follow.

That drives me absolutely crazy because that was one way that political crap that I didn't want got shown to me.

Also, this is shocking I know, but I don't agree with or like everything that people I know and/or follow think or like. So a lot of these posts were from pages that I am already familiar with, have already decided I don't like and want to not follow, and yet it STILL gets shoved in my face because people & pages I follow are into something that they are objectively wrong to be into and should thus rethink every life decision they ever made over the years that helped to drive them towards being interested in such garbage.

I know I'm old but all I want in any "news feed" is to show me latest posts from people/pages I follow directly in chronological order with pagination and no automatic refreshing of the news feed etc.

But I guess I'm not the type of user they are chasing .. and I find myself organically being less and less interested in engaging as a result. I don't need to break social media addiction ... the social media "platforms" do that for me with their "features."

That is the "For You". I have a comment on this story about how you can remove it, and there are plugins out there that remove it.

Once it is gone Twitter becomes usable again – you only see who you follow and ads. Otherwise it is a really awful experience.

> I know I'm old but all I want in any "news feed" is to show me latest posts from people/pages I follow directly in chronological order

It sou ds to me that you want Mastodon

You can disable "For You" and only have your timeline and lists.

But you need a plugin (or run some javascript). I am trying to get a Safari Extension through the App Store... but that is taking some time.

    (function() {
        var defaultMatch = "For you";
        
        function fixYou() {
            let firstSelected = document.querySelectorAll('[href="/home"][aria-selected="true"]')[0];
            let text = firstSelected.textContent;
            
            if(text == defaultMatch){
                document.querySelectorAll('[href="/home"][aria-selected="false"]')[0].click();
            }
        }

        const observer = new MutationObserver((mutations) => { fixYou() });
        observer.observe(document.body, { childList: true, subtree: true });
        window.addEventListener('load', () => { fixYou(); });
    })();

This is the code to do it, if you want to roll your own.

I blocked it via uBlock origin

They have done this since forever, even under the previous ownership. The “flavor” of the political content was different then, but it’s always been annoying. One of the worst features of the site.

Selling political engagement seems pretty desirable if you are a platform unencumbered by conscience. Unlike products, there’s no way to close the loop.

If you are trying to sell engagement with, say, some product on Amazon, presumably the seller will eventually wonder why nobody is clicking your tracker links and buying their products. But the targets of political ads get to go vote in private, nobody knows how effective the propaganda was!

Any media pusher that decides the ranking of stuff will have the same bias in some direction. Only thing I can think of that could counter-act that is placing the ability to rank things at the hands of users, like Bluesky with the Feeds feature, or Mastodon that just flat out refuses to rank content at all.

At least then users can pick their own poison, rather than be force-fed one.

Gosh you're totally right, the article has pitched this as a problem unique to Twitter, but of course Facebook and Tiktok et al have the same thing going on, with minor variations on the theme.

TikTok was always full-throated, immediate engagement fishing. You watched a video on [some thing] for 30 seconds, you were going to get fed a bunch of that right after.

Twitter used to soften the edge on this a bit; you engaged with a few [some thing] posts and you'd get 1 or 2 more over the next day. Now you can very quickly get into an infinite loop. The "algorithm" (and yes, I hate writing this) digs its claws in right after an engagement.

Facebook (for all it’s many problem) seems to have figured out that I don’t want to see politics. I think it is just generally not a site that people go to for that sort of thing.

Facebook shows me a lot of politics, some of it I find repugnant and much of that is clearly designed to troll. Interesting that your experience is so different.

FB also shows me a lo of pseudo-science, similar history, etc. because of my interests in the real things. It has recently taken to showing me pictures of Eastern Europe for no apparent reason.

I get some pseudo-science (but it is funny, ancient aliens and quantum bullshit), a little bit of politics (but like, links to the Atlantic or whatever, so nothing too offensive), and lots of gaming stuff.

Overall, I really do not like Facebook (I’m only on there because some relatives are). I just think their algorithm is somewhat effective. I think I’ve managed to get them to label me as somebody with strong political opinions but low emotional valence toward politics, by marking some political posts or pseudo-political posts as offensive to me, but never allowing political content to “dwell” in premium screen space in the app.

Edit: I say I think their algorithm is effective, although I should say, I’m pretty sure I did trick it. In fact I’m very concerned about politics but I make sure to only angstily read articles I disagree with in Firefox browser with lots of tracking protection on.

> Overall, I really do not like Facebook (I’m only on there because some relatives are)

Similar to me. Friends and relatives. Its also how the home education community in the UK communicates (I admin two groups, and that is my main reason for being on it regularly), an unfortunate reflection of the hold FB has on the demographic (middle aged because they are parents, mostly women because even in the 21st century men do not bring up kids).

> Gosh you're totally right, the article has pitched this as a problem unique to Twitter, but of course Facebook and Tiktok et al (...)

Why do you think this is relevant, if the subject is Twitter's algorithm?

Your comment reads like whataboutism.

This is an honest question for people still using X/Twitter and yet complaining about it: why?

It's been obviously going downhill for a while and the reason has been obvious too. The leadership doesn't care about your concerns.

Why not stop using it or move to something else?

There are a bunch of alternatives out there and most people don't have a job that requires them to specifically use Twitter.

Unlike Facebook, where I know I have missed out on communities by not using it, I don't feel like I've ever lost anything by not using Twitter. Anything that goes viral on Twitter these days gets a bunch of mainstream news articles anyway.

> This is an honest question for people still using X/Twitter and yet complaining about it: why?

I agree. Just stop using it already. What's the value of subscribing to a Kremlin bot sandbox? Sometimes these complaints just sound like attempts to keep people engaged after clearly identifying the elephant in the room.

> Unlike Facebook, where I know I have missed out on communities by not using it, I don't feel like I've ever lost anything by not using Twitter. Anything that goes viral on Twitter these days gets a bunch of mainstream news articles anyway.

The residual reason to use twitter isn't stuff that goes viral on Twitter, it is an existing network of specific individuals on Twitter (which ones are different for each user for whom it is an issue) who haven't moved off.

For anything AI news it beats the other platforms by a mile.

I don't see the difference between it and Mastodon or even Reddit in that regard.

Use Twitter lists to bypass the algorithm:

1. Create a list of accounts you want to follow at https://x.com/<your_username>/lists

2. Bookmark your list instead of the main feed

3. Optional: Use a browser extension to auto-redirect to your list

Result: A chronological feed without algorithmic interference.

For the past couple weeks or months, twitter has been shuffling the order of sidebar buttons to make room for new features e.g "Premium", "Grok", "Verified Orgs" "Jobs" "Communities"... with no way of disabling this helpful 'feature' or changing the order. "Lists" is now hidden in the same submenu as settings. Way to kill my muscle memory...

How many people (percentage wise) use a browser vs app on mobile? Seems like accessing via browser would be much smaller, and I'm honestly surprised that they haven't crippled its use to "encourage" the use of the app.

That's great once you already have a list of interesting accounts you want to follow, but it doesn't solve the account discovery problem.

Start with a trusted account (eg qntm) and add some proportion of who they retweet to the list. Do this recursively.

That's exactly how I do it.

I am very interested in politics and political science but I am not at all interested in what people have to say about politics on Twitter, Mastodon and such. There is some value in probability polling the "silent majority" and in serious reporting but a self-selected sample can send each other image memes all day and it doesn't matter.

On Mastodon I post articles about politics in France, Brazil, Nepal and such and also about political science but I have rules that block words like "fascist" as well as the names of major US politicians -- I don't care what tribe people are in and the more you make me know the more I go down the Unfollow -> Mute -> Block path.

Bluesky's algorithmic feed has a bias towards nice and the people may be nicer too (e.g. Bluesky has "chronic illness" cases to but unlike on Mastodon they don't blame me) and I see a lot less anger than I see on Mastodon without having to set up a lot of filters. Still I am doing the "follow a lot of people to get some to follow you back" thing and finding it is emotionally draining to scan through a few hundred profiles and reject all the ones that talk about politics. It's a job for my agent YOShInOn.

It's one week before the election, it's hard to avoid election content no matter what platform you are on.

A great use of its Grok AI would be to autotag posts as "political", "sports", "crypto", "erotic", etc. It probably won't be perfect, but avoiding whole topics would become much easier.

Maybe that would be expensive and harmful to engagement, but one can dream.

Not that a LLM wouldn't shine at this, but topic classification is a fairly well-solved problem with smaller, easier, cheaper ML approaches.

So does every social media website except for niche forums.

Reddit has transitioned into 100% propaganda mode. Every single post and comment is political astroturfing.

Must admit I haven't noticed this yet in /r/emacs.

different kinda politics there.

Yeah, it's even invaded the more obscure subreddits. Some posters try to shoehorn in political commentary even when it doesn't have much to do with the topic. I doubt there are even real humans behind some posts.

> I doubt there are even real humans behind some posts.

The topic of Russian bots pushing all sort of propaganda designed to destabilize western countries is a well known problem. There are already multiple discussions on stunts like blatant sock puppet accounts and poorly-built LLM-powered bots dumping massive volumes of propaganda.

This problem is well established. Until now the answer to this problem was relying on social networks to moderate abusive content and actively fight state actors from totalitarian regimes engaged in what is ultimately warfare against democratic nations. What makes Twitter stand out is how one of these totalitarian regimes ended up compromising a major social network and is using it's leverage to attack other nations with free reign. This is what makes Twitter stand out.

I'd wager it is almost certainly real humans, just not engaging authentically i.e. paid trolls, at least until the cost of generative AI falls bellow troll farms.

If you bother to spend a couple of minutes looking for it, you'll quickly stumble upon a high volume of posts unmasking LLM-powered bots, which even respond to prompts to clear previous prompts, output their own prompts, and even generate content on-demand.

> I doubt there are even real humans behind some posts.

I doubt there are humans behind as much as 40% of some subs.

I have an account I exclusively use to follow users of a certain topic. I don’t really get political content, even when using the for you tab. That being said I’d love for there to be an option to ditch political content on my main account.

Same here.

My Twitter is purely construction technology (thrilling I know) and I never see anything else.

Also they stopped resetting me back to For You, so I only see content from accounts I actively follow or they retweet.

I would love to see more in-depth analysis of recommender algorithms. You don’t need to know how they work (on a technical level). What matters is the results they are showing to people.

In lieu of an FDA for algorithms that I would love to see, it would be great to have some basic assessments of the distributions of recommendations an algorithm is making.

A few years back when he seemed at the height of his popularity, my colleagues built a little tool to check how many YouTube suggestions you had to click to end up with Jordan Peterson.

IIRC, he was never far away from anything related to history, politics or philosophy; but hard to reach from anything else like sport or music.

Write-up here: https://eticlab.co.uk/jordan-petersons-proliferation/

This totally aligns with my personal experience. I am Italian, most of my "For you" tweets used to be in Italian. Recently a lot of US, pro-GOP content sneaked in; stuff along the lines of: guy explaining that his wife must vote as he wish (i.e. Trump); anti-abortion lady praising Musk; anything (good or bad) mentioning Musk. I am sure I am doing something wrong on my side and started falling to rage-bait, but I am equally sure I didn't initiate any interest in this kind of content.

I don't have an account and just redirect x/twitter links to xcancel so that I can still see comment threads. It also provides RSS feeds but I haven't tried using them. Doesn't solve the account discovery issue but I think my sanity is better off that way.

True - you really can't escape from politics on x. It's really just a rage baiting.

I'd say this is pretty much true of Threads as well. I had a few months experimenting with it, turning down (I'm not interested in/outright blocking) any political stuff that came my way and it still filtered through somehow.

How can an algorithm determine whether content is political or not? Sure you can naive filter for a list of political urls but it won't catch most of them.

On one hand, we have HNers telling us that AGI is already here.

On the other hand, we think that we can't effectively classify the majority of posts.

On the third hand we have a VC backed startup selling AI content filtering with a backdoor advertising deal.

You don't even need GPT-2 level AI to classify tweets.

yeah simple sentiment analysis is a basic 4th year undergrad / master's level project.

can be easily done by one person, and with a fairly good level of accuracy and key-word tagging. emphasis on "fairly good", but diminishing returns and all that.

> How can an algorithm determine whether content is political or not?

For starters, cross-referencing it with popular propaganda talking points. Smear campaigns are a dead giveaway, and adding two or three topics should be enough to cover the majority of propaganda posts.

I'd argue that just having an effective anti-bot policy is enough to render this a non-issue, but apparently Elon Musk prefers Twitter to be a Kremlin bot sandbox.

Yeah, must be a coincidence that people are seeing a lot of alt-right stuff!

You need to use the block option, relentlessly. There's loads of garbage on there but you can clear much of it away. See it -> block it -> done.

level of effort vs. payoff. my interest level in a lot of twitter posts is very, very low, and blocking constantly just isn't worth my time. at some point you just have to concede the platform is dying and move elsewhere.

In most parts of the World, some aspect of the media is regulated.

In the UK, broadcast news is heavily regulated by an independent body (Ofcom), and newspapers are "self-regulated" by a body created after the phone-hacking scandal[1] staffed by newspaper insiders (IPSO). Even so, sometimes bad things happen and Ofcom has to take GB News to task, and IPSO has to tell one of its mates off for printing something invasive. This is all below the level of the law coming into play (for example, slander/libel or personal data handling), this is just "best practice".

The regulators are needed in the UK and everywhere else in the World they exist not to censor platform owners and contributors, but to protect the wider World from them. You can't just break into somebody's voicemail and report what you heard any more. You can't just have one politician interview another and pretend its news content, no matter what GB News say. There are standards to try and improve the quality of the output to make it useful.

Social media has clearly reached a point where some rules are necessary for that purpose. Meta have had a little swing at this with their own Oversight Board, but its clearly not working enough in the interests of the audience because we keep finding examples of dead teenagers whose parents insist could have been saved if Meta had just been better members of society. [2]

And then there is Musk. I get it, he started a company that builds cool rockets, and bought Tesla and drove it bigger, and he had a cameo in Iron Man 2 because Tony Stark is meant to be modeled on him by the film creators, but it's clear now, he can't run an influential platform well, and it's no surprise that even if he isn't actively promoting his strong political preferences in the algorithm, he's not exactly diving in and insisting it gets fixed to solve this problem.

Regulation is seen by the tech industry as an impediment to innovation, a barrier to growth. I think with technologies that impact the health of a society, that society has a right - a need - to see that those technologies are regulated in the name of the society not being harmed by it.

[1] https://en.wikipedia.org/wiki/News_International_phone_hacki...

[2] https://news.sky.com/story/molly-russell-inquest-teenager-di...

https://edition.cnn.com/2022/04/19/tech/social-media-lawsuit...

How about regulating teen accounts? If you want to use it, your parents must sponsor you. Then if you blame FB or whatever for the teen's suicide, FB can pull the logs and show how much interaction the parents had with their child? If the parents appear to be derelict, then let the court assume that neglect was the main factor.

> The regulators are needed in the UK and everywhere else in the World they exist not to censor platform owners and contributors

You say that, but then again:

https://en.wikipedia.org/wiki/Give_Ireland_Back_to_the_Irish...

https://en.wikipedia.org/wiki/Greece:_The_Hidden_War

https://en.wikipedia.org/wiki/Al_Jazeera_bombing_memo#Offici...

First one is irrelevant: McCartney refused to deny he was fund-raising for a terrorist organisation that was literally bombing the UK at the time. That, and the fact the majority of people living in Ulster wanted to remain part of the UK (and still do, but demographic changes mean that might not be the case in a generation), meant it was a slam dunk that the BBC would ban it to prevent offence that could lead to an incitement of violence at a delicate time. Also note that wasn't a regulator taking action, it was not the government, it was a broadcaster coming up to a charter renewal conversation. Complain about Auntie BBC if you must, but not a fault of regulators.

The second one is irrelevant in the modern context: the IBA hasn't been a regulator for over 30 years, and had a different set of guidelines and authority to Ofcom. It is unlikely that a fair and balanced account where a reasonable right to reply was offered to all parties would be banned from broadcast today.

The third one is a ban under OSA, and has nothing to do with media regulators, its about the scope of acts like the OSA, and its appropriate use.

What do you think of the obvious bias of the New York Times?

The answer I expected! :-)

Why is Twitter only a problem to be solved now? Twitter was arguably more biased (in the opposite direction) before Musk took it over?

or maybe also everything has become politicized

I don’t buy this. Bluesky serves me multiple feeds with no politics in them. I have a game dev art feed, art, gardening, none of these have politics. I think the more likely answer is that politics is more likely to cause doom scrolling, and thus, ad impressions.

What I’m saying is not offering algorithmic choice is, in itself, a choice that X makes.

Elon Musk is also very heavily "invested" in the outcome of this election, if you believe he might be putting his thumb on the scale of the algorithm to his own ends. It wouldn't surprise me after his own account started showing up constantly on everyone's feed not long after he took over, I had to block him to get him to go away.

The first paragraph of the article contradicts this notion. I go on instagram and don't get much if anything directly political about the US, let alone Jan 6th bullshit. I get stuff on Gaza but its from one account that is a friend who I have not muted or unfollowed.

Agree. I will be glad when the longest election cycle ever finally ends

There is a huge button called "following" when you press it only the feed of people you follow shows up.

It's too bad that you have to take an additional step to do something that ought to be the default behaviour.

It stays default if you don't switch to For you tab.

And when exploring for new interesting accounts to follow, that huge button is useless.

So does the frontpage of WSJ, though of course it leans in the other direction.

Ability to block certain topics completely would be nice -- though it goes for all media. With ever smaller / faster (local) LLMs, I could imagine a nifty browser extension that does that.

> though of course it leans in the other direction.

Besides what the others have noted, are you implying that the Twitter is leftist, or that the WSJ is?

Overall, both the WSJ and Twitter very much lean right in the political spectrum.

Yeah you're being downvoted but I think OP doesn't know what WSJ is or which way they lean and is just sniping at the existence (?) of journalists.

Just use cosmetic filtering on ublock to filter for certain words using has-text in [span, a, p] elements (div breaks some websites, thus doesn't work for hn comments). It helps remove a lot of noise on the web. Particularly on select topics that get astroturfed / rage-baited to hell.

e.g.

##p:has-text(/tiktok|tik-tok|twitter/i)

Do you think the Wall Street Journal and Twitter are functionally the same?

The Wall Street Journal is a newspaper.

> So does the frontpage of WSJ (...)

Apples and oranges.

Reading the WSJ is a kin to subscribing to a specific Twitter user: you know what content to expect, and you are interested to read more.

Being force-fed political content you explicitly have not subscribed to is like you're reading the WSJ and out of nowhere you get pages from, say, MAGA's newsletter.

Not the same, is it?

Heh, so do the editorial feeds of newspapers. I think that if political content is allowed at all, users should be given the full gamut. Not a feed tailored to their beliefs (either affirmative or ragebait), because echo chambers are radicalization machines.