Hacker News new | past | comments | ask | show | jobs | submit login
Twitter launches its ‘Hide Replies’ feature in the US and Japan (techcrunch.com)
118 points by antibland 20 days ago | hide | past | web | favorite | 107 comments



This is to me seems like a very counterintuitive “feature” — at least on paper. Hiding tweets seems akin to telling the whole world, “here are tweets from my critics I dislike so much I’m going out of my way to suppress them”. At least when you report, block, or mute someone, no one else knows that you were bothered (or, as trolls like to say, “triggered”).

It’s not the same like being downvoted on HN or reddit, where the vote to suppress is a community/crowd decision. If Taylor Swift hides a tweet, you are inherently curious to know what rando’s tweet managed to hurt or anger Swift (or her social media minder).

That said, while Twitter seems to be incredibly slow at introducing new features, the trade off is that what they do push out often ends up being a good thing. The survey evidence cited in the article seems flimsy, but I give Twitter some benefit of the doubt that they wouldn’t push an obviously destructive feature.


> That said, while Twitter seems to be incredibly slow at introducing new features, the trade off is that what they do push out often ends up being a good thing.

I find most new Twitter features to be worse than useless:

• "Like" was turned into "algorithmic retweet". Now I can't like a tweet without some percentage of my followers seeing it. This is also a problem when I follow others. I can disable seeing their retweets but I can't disable seeing their likes. I've had several people tell me that they unfollowed me because of tweets I liked. I've done the same to others.

• My timeline now shows tweets from people I don't follow. Above the tweet is something like, "Tim Urban and 8 others follow...", and then there will be a tweet by someone who I really don't want to follow (such as Ben Shapiro). There seems to be no way to disable this feature.

• Quote retweet is a great way to encourage flamewars. Long ago, people could not retweet with a message. That meant that people were less likely to retweet something they found inflammatory because their followers might think they agreed with it. Quote retweet means they can preface it with, "Look at what this terrible person said." That's a great way to encourage the Toxoplasma of Rage[1].

Overall, the effect of these features has been to reduce my Twitter usage. Now I mostly go there for a couple of group DMs.

1. https://slatestarcodex.com/2014/12/17/the-toxoplasma-of-rage...


Twitter is an interesting case where I feel that they found their ideal feature set by around 2011 or so. They had chronological tweets, likes and retweets worked logically, everything was great. Unfortunately, they never made any money this way and needed to increase "user engagement" to show us more ads. So they abandoned the chronological timeline for what they thought was best, they started randomly showing stranger's likes. At least they've finally given back the option to switch to "Latest Tweets", but it's pretty telling that the site 'forgets' that setting every month or so and tries to show you what they think you want again.


> "Like" was turned into "algorithmic retweet". Now I can't like a tweet without some percentage of my followers seeing it. This is also a problem when I follow others. I can disable seeing their retweets but I can't disable seeing their likes. I've had several people tell me that they unfollowed me because of tweets I liked. I've done the same to others.

G+ had this too, and I also got threats of being unfollowed if I didn't stop liking things. (I could turn it off in my settings, though, but I liked to use reshares to add comments and +1 to reshare without comment.)

Ultimately what the social networks never delivered on were the true concept of areas of interest. You should be able to follow someone's interest in a certain subject, and then like things in that context. That way if someone is following you for your programming content, they don't have to read the anime reviews that you like, or whatever.

It never happened. I think social media is a dead thing now. You can use it to snipe at political figures and promote products for money. That's about it.


I was going to mention the G+ usage.

I really disliked this for numerous reasons:

- It was an overloading of the +1 button. Now a "like" was also a "here's my plus-one, repost me, maybe?".

- The behaviour wasn't obvious to the upvoting (and re-posting) user.

- People will and do upvote stuff they'd never re-post.

- It was seriously annoying.

I would suggest people disable the feature, if I noted it, or I'd unfollow them. There was no way to restrict such reposts from my own stream, other than (as I eventually discovered) avoiding the home stream entirely and viewing entirely by Circles. I had a set of higher and lower-priority Circles I'd follow, which actually helped cull most of the crud.

I strongly suspect the design decision was based on the fact that very few people actually post or repost material. They read on a consume-only mode, if at all. Google Plus's active user base was a small fraction of 1% of all profiles, and even within that set there was a phenomenal range of activity. The effect being that most of the platform was dead air.

My read is that while the extent of this may have been more so at G+ than with other leading social networks (Facebook, Twitter, Reddit, etc.), the general principles aren't. Power laws and Zipf functions are extraordinarily prevasive, and show up virtually everywhere. The problem is that there's a very narrow band of presence between "shut the heck up already" and "I thought you were dead". In the real world, distance and/or proximity mediate this. Online, where everyone is (very nearly) as digitally close to everyone, we need other means of damping the overexuberant and enticing the timid.

The "areas of interest" concept finally, sort of, got fleshed out at G+ by way of Collections, though even that was pretty butchered. Topic/channel discussions (as with Reddit / Usenet) still seem hard to beat.


> You can use it to snipe at political figures and promote products for money.

I think that's wrong - there are lots of communities of interest where people follow a crowd of others because of a common topic of interest.


FB Groups, subreddits, discords all fulfilling this purpose. The thing with these products is you seem to either get “social graph” focused or “interest” focused.


I hate those first two a lot. It has definitely caused my Twitter usage to decrease. What metric could Twitter possibly be optimizing for, and why not at least give people the opportunity to disable them?


  > I can disable seeing their retweets but I can't disable seeing their likes.
You can stop seeing "likes" (in general, not from a specific person) if from the official client you mark the tweet as "I don't want to see this" (I don't know the exact message in English) and after giving this negative feedback several times about likes, the client stops showing them.


I do this whenever I see those, but after two or three weeks, the app will again start shoving this garbage in your face. Rinse and repeat.


Use the "Latest Tweets" mode instead of Home, you can switch from the stars icon in the top right.

Latest tweets doesn't show likes for me, ever. That, and also clicking "I don't like this" on a bunch of other algorithmic suggestion boxes, means I'm seeing just the regular old timeline of what people [re]tweeted...


Same here


For 1 and 2, simply set your timeline to "latest tweets".

For 3, sometimes people simply need to be made fun of. The QRT is a great way of achieving this.


Setting my timeline to “Latest Tweets” works only for that session. So add that to the list of shitty “features” that the OP peeled off — the fact that I now have to take action to use Twitter as a chronilogical feed.


I've had it be persistent ever since it was introduced.


Obviously 3 is effective at what it does but it's basically the definition of putting someone on blast in front of millions of people. It's bullying on a supermassive scale.

- For every bad person that's had their terrible opinions aired in public there's a right wing talk-show host that users RTs to ruin the lives some small time LGBT podcaster or youtuber with their fanbase.

- If the person is big enough to take the heat you've just given them an even bigger platform for their ideas.

- In general I don't think anyone deserves this even if their views are shitty.


> - For every bad person that's had their terrible opinions aired in public there's a right wing talk-show host that users RTs to ruin the lives some small time LGBT podcaster or youtuber with their fanbase.

I agree with using it for the first, but disagree with using it for the second, and as such those people should be banned! A rather simple solution.


I bet it'll work.

First, I think most of the time the hidden content is going to be very low-quality (and thus not gratifying to explore), rather than "juicy", and users will quickly learn that.

Second, it puts a huge damper on the gratification of stirring sh#t and getting a bunch of activity, notifications, and subs as a reward. Instead, when you stir the pot, you find yourself ushered to a closet by yourself. It's not locked, but you're definitely not the life of the party anymore.

The result is a social force– an adjustment to incentives— that keeps conversations on-topic and civil. Really, I think it's missing the point to frame it as a tool for silencing critics with dissenting opinions, instead of a tool for silencing people who want to call you an ${inflammatory_statement} in your thread about a cake you baked today.

As I understand it, as your follower count goes up on Twitter, the latter problem gets almost untenable to handle because the strategy of coat-tailing high visibility accounts by sh#t-stirring in their threads is currently so unambiguously effective.

Time will tell what effect this has, but I am hopeful that it will have a large positive impact. (I think the problem demands action no less decisive and significant than this).


> The result is a social force– an adjustment to incentives— that keeps conversations on-topic and civil. Really, I think it's missing the point to frame it as a tool for silencing critics with dissenting opinions, instead of a tool for silencing people who want to call you an ${inflammatory_statement} in your thread about a cake you baked today.

If that's really what it is, let's disable it for accounts representing government entities and elected officials. Seem like a good compromise?


There's legal precedent that public figures blocking constituents on social media amounts to a first amendment violation, so that idea may actually hold water. Naturally that only applies for government officials; the first amendment doesn't (and shouldn't) prohibit private entities from curating content.

Although bad-faith dialogue disruption does happen with political and government officials too, and can be a problem (including potentially a free-speech problem, for those who are drowned out) for different reasons.


> There's legal precedent that public figures blocking constituents on social media amounts to a first amendment violation

That's US-only.


There's also the 'Snowden Problem' which is contractors that act on behalf of the feds don't have the same requirements and protections as the feds.

And around 95% (my assessment) of 'federal employees' are contractors.


I'm coming at this from the other side, but as you note, it'll be very interesting to see how it plays out. As a consumer, I don't care what the original poster thinks I should see - they may not even be someone I follow if I click a retweet for example, and if anything, seeing what someone wants to hide is much more interesting. It's a net negative from my end because now I have to click more to see the actual content but I am far to cynical to assume that's anything more than an attempt to drive engagement metrics!


Do you think it can be used as some sort of metric? Like ratio of followers:blocked replies? Or the number of accounts which block a person's replies?


One thing I've noticed is that getting many Likes or Retweets on a single tweet rarely leads to increased follows.


Given the number of steps required to hide a single response, if a tweet ends up with 50 abusive responses, whether genuine or bot generated, who is going to put the effort to hide them? That’s not going to stop twitter from being anything else than the dumpster fire that it is now.


If they are smart, twitter will make that option available in the API so that social-media managers can do it en-masse from some sort of 3rd-party power-user dashboard.


I think that this is kind of a good thing. If you abuse the "hide tweets" feature to block people giving fair criticism, you'll probably get called out on it. But if you use it only to hide true garbage replies, then no one will care. The fact that other people can see what you've hidden adds a sort of accountability to the system, which to me seems like potentially a good thing.

It seems like a good compromise between entirely open discussion (where people can troll without consequence) and a Facebook-like system (where people can silence all dissent on their page). But who knows -- we'll have to see how it actually works in practice.


Hide truly garbage replies first and no one will scroll past them. SMM will generate lots of garbage to themselves to bury real damage under kilometers of junk. “Sorry for that, we recently managed an attack by unfair competitors.”


well that's interesting, I haven't ever really seen SWIM used in a context outside of generally drug related forums and discussion


Wouldn't you just hide the criticism too though. Considering the worry around filter bubbles, elitism this be one of the strongest ones of all?


That's a good point. I'm only thinking in terms of Twitter celebrities (both the real-life and only-on-Twitter variety). But for more average users who regularly tweet and discuss controversial (or otherwise troll-attracting) topics, and get a moderate number of good and trash replies, this feature would likely be a good quality-of-life improvement for their followers.


Yes, I say that happen.

An account reposted a tweet as their own and then hid all the Replies flagging it as stolen content.


People will use it differently. If you use it to tune out noise, nobody will bother to check in yours. If you use it to quell rational dissent, then it will be the go to place for the juice.


> At least when you report, block, or mute someone, no one else knows that you were bothered (or, as trolls like to say, “triggered”).

That’s not really anonymous. For example when someone reports a tweet, Twitter also sends a message to the concerned person to let them know someone reported them, even if they didn’t take any action. In case of cyberbullying, the abusers can use the Twitter report as a tool to add more drama and pour oil onto the fire.


Seems like a good move. Better than deleting.

First thing I do when I see "more replies" is click on it, knowing this is where the juicy stuff or counter argument is.

Like in HN, when it says [dead], I have to read it, to see whether I agree with it being [dead] or not ;-)

So there is certainly psychology there, at least to me, that makes me dig further. By hiding stuff behind a click you actually make me want to read it more!


> First thing I do when I see "more replies" is click on it, knowing this is where the juicy stuff or counter argument is.

Exactly, which is why I don't understand this feature. If you can't really hide replies, only instead call attention to the replies you've "hidden", what is the point of it?


Already seeing people screenshot hidden replies and tweeting with "gee I wonder why this person hid this reply lol" type messages.

I'm skeptical this will have the civilizing effect they hoped for.


Same thing happens with blocks!

But the way the feature works is to fork the conversation. Now, both ideologies can continue their threads of discussion without necessarily disrupting the other one.

I always think of kuro5hin when these types of containment features are brought up. It's kind of like "your freedom to swing your fist stops right before my nose" type of protection.


Hm, yeah it seems like it would make more sense to have something like "Comments are disabled for this video" on youtube.


It would be weird for twitter though because a tweet itself is essentially a comment- people would make de facto counter-comments instead of replies and it would just be an unthreaded version of the original thing.


> Like in HN, when it says [dead], I have to read it, to see whether I agree with it being [dead] or not ;-)

And if you don't, vouch for it :)


"I Disapprove of What You Say, But I Will Vouch to the Death Your Right to Say It"


How about "I disapprove what you say, but you can go say it on your account, and not in mine, because your replies will be hidden, because it is my account.".


I did that — vouched for comments that I thought were on-topic and dead only because people (sometimes including me) disagreed — and promptly lost the ability to vouch.


Trump will hide all the tweets against him. Not sure how this is a good feature.


There's a lot more to trump than twitter, and a lot of discussions have been ruined through trolling, both by individuals and bots. This provides a way for people to clear their own front yards of what they consider trash, but there's an abundance of yards. If Trump hides all dissent, so what? It will be more than apparent, and right behind a single click!


Argh, that should be "there's more to Twitter than Trump," not the reverse.


This might go down the wrong way in India. We see a lot of misinformation, propaganda and often messages inciting mob violence floating as tweets. Reporting these tweets does not do anything immediately, if ever it does anything at all.

We used to post a reply linking an article with correct facts and some context. See https://www.altnews.in/

This feature would make this fight a lot harder.


What about when the misinformation/propaganda/etc. is in a reply instead of in an original tweet? Two-sided sword, as they say.


The first time I saw this in the wild, the author had hidden a single tweet mentioning that they had received 50k retweets on a stolen video.


Russian state propagandists are going to love this feature. Right know they're constantly fact checked and called out on their hypocrisy in the replies but soon they'll be able to peddle their "death to Navalny/Ukraine/Europe/America/homosexuals" rhetoric unchecked.


Politics about any specific country or group aside, this would be a win for Twitter. Extra engagement from both sides. People who write crazy things return to moderate comments. People that comment write more comments because old comments are hidden. After comments are hidden, someone will probably start their own tweet about how that other tweet is full of garbage.

It's just more gasoline for twitter's business model: Drive engagement through flamewars.


ALL propaganda agencies are going to love this. State or industry. Pepsi or Joe Biden.


I don't see a single person in the top comments speak about this. This feature is going to allow people to hide legitimate non-trash responses that criticize the tweet, and it will be very convenient going into the 2020 elections. Dissenting voices WILL be silenced.


"Prepare for unforseen consequences."


Won't this just make Twitter even more of an echo chamber?


I can't imagine why anybody would consider this controversial in the least. Id we are living in an attention economy where anybody is allowed to make a small debit from your account at will. Giving people more tools to manage those debits should be lauded.


Kinda like a website that is limited to 160 characters. Never fell in love with Twitter and looking at the top trends I often feel like wasting time on that platform.

But I agree that Twitter is basically an attention distributor. I get the impression that Twitter focuses its ambition to pander to self-expression and many topics that could have been interesting degrade to something that feels even worse than cheap PR.

Sadly, however that happened, Twitter became a platform to advertise political positions. That is probably why so many people see such features with mixed feelings.

I disagree that people make debits by replying, even if it is done in bad faith. If you put something out there, you are likely doing that yourself and nobody takes anything away. By hiding something you take something away from readers.

#iAmAgainstEverythingThatHappensOnTwitter


What would you think of spam? Isn't that clearly an attention debit? I would say that all replies, wanted and unwanted, take at least a small amount of attention. And if the fraction of useful to not useful replies decreases, that's the SNR going to crap.


I believe Jack Dorsey when he says he wants to improve the quality of discourse on the platform, but I'm not sure this step will help anything. If people are being harassed, Twitter should deal with harassment. Adding the ability to hide replies simply allows people to handwave away discussion or criticism that they don't like, which seems detrimental to rational and reasoned discourse.


Seems like a reasonable response to the criticisms, but also, I dislike how Twitter just gradually gets closer to Facebook every year. I don’t like much social media in general, but I kind of liked the old, very limited Twitter. It had a zen-like property to it, in some sense.


I read this as hide reptiles and got very excited. I would love to have a way to not see photographs of reptiles, especially snakes on the internet


As a lizard, I find that offensive! ;-)


At least you're not a snake!


twiiter is going into history as the utmost failure of human communication. How such a crippled way of communication (limited characters and structure) combined with such childish disregard to discourse (blocking and shadow banning) could form a platform for "discussions" of important topics like social change and politics, I'll never know.


The reason Twitter became big was coincidental more than anything else. It became the defacto tool used in the Iranian protests against Ahmedinejad in summer 2009 (and again in 2011 during the ME-wide uprisings) as the regime had blocked Facebook to keep a lid on international awareness. The use of Twitter gathered a ton of foreign media attention as there was no other source of primary information.

And almost overnight, Twitter went from an obscure "microblog" (nobody's used that term in years) to "the pulse of what's happening now".

The cable news media (and digital media startups a few years later) were a willing partner in its rise. Embedding people's tweets in their stories gave them the ability to generate "news" to fill their 24-hour programming at a fraction of the cost of doing investigative journalism. The latter is by its nature lengthy, wrapped in secrecy and doesn't always result in a scoop. Basically all the things that a ratings and ad-driven industry couldn't care less about.


Oh dear, where will I look now to see the same seven reaction gifs forced into bad jokes every single day on every single post?


When people started using emoji to express themselves, I thought it couldn't get any worse... then came the reaction gif.


Just browse giphy. But be careful. You never know what you'll find there.


I dunno. Just dont use twitter? I never had an account and never saw the need. No instagram or snapchat. only thing is a mostly dead facebook page that rots. I'm doing just fine.

I just don't get any of these social media is ruining my/our life when you are the one who opts in. It's like a person who complains the fire they stuck their hand into burns and then continue to keep said hand in the fire.

Want to be social? Go find actual groups of humans who have similar interests. Shocking concept, I know.


I've never been into twitter either. There's a lot of content there though, I'll visit the site from time to time if another site links to it, but that's pretty much it.

Politics on twitter is just a shouting match. Most tech related things are just kind of link-sharing (sometimes people post lengthy bits of tech info, I can't understand why). Twitter does seem to be a decent venue for breaking sports news, but that news usually finds its way elsewhere quickly.


You should hop off hackernews then and find some grey beards to hang with


Hmmm. One of the reasons I have a Twitter account is to call out shitty companies (such as my phone company) when they basically ignore my complaints via email or phone call. A bit of public shaming seems to work wonders in these cases.

Now I am going to have this option taken away?


If you comment on their tweets, yes. If you just tweet something out, Comcast can't hide it (yet).


Reading the name I thought it be a way you could directly reply to a tweet without it being such a public matter, and also gunking up my timelinem but it seems to be a way to self-moderate your own posts which is... interesting.


It seems like using an upvote/downvote system akin to Reddit and Hacker News would make more sense for this goal. Allow all viewers to determine the quality of a comment, not just the person who posted the original tweet.

Everything we've observed the last few years shows that we should be democratizing discourse on social media, not further funneling the power of the conversation into fewer hands. Do you really want extremist twitter accounts to be able to hide all dissenting replies?


What on Earth is wrong with techcrunch? I clicked the link of the story, read it, decided to go back to HN. I had to click the back button 4(!) times.


The url changes a few times while scrolling.


I’d like a comprise feature that automatically paired back the annoying bots, such as only allow my followers and my follower’s followers reply. That way you get enough of a network effect to get interesting replies without the garbage.


Yasssss. This is the future, no replies. Don't @ me.

EDIT: After reading more about this, it looks like people can still reply? Would be nice if they just added a straight up "Don't @ Me" feature where you can disable comments completely.


Would be semi-impossible given that half the comments shown on a Twitter “thread” aren’t really comment-objects, but rather just their own tweets sitting on someone else’s timeline that have @ed you in them and twitter figured out which thread to attach them to by heuristic. Sure, in theory, you could turn off the heuristic—but you’d still see them just fine if you did a global search for your @handle.

Or are you suggesting disallowing what is essentially “allowing people to refer to you on their own blogs”?


>in the US and Japan

Why only these countries? I also see this with Facebook Dating; it's only available in a few countries right now. Why is this? It's software, can't you deploy the feature everywhere at once?


Trial-marketing. Twitter users generally tend to form relationship (follow/see/RT)-graphs stick within language boundaries (and to a lesser-extent, national identity boundaries) - by using language or region as a proxy for a "clean" subset of users you can get some experimental results that you won't get by using random user selection, for example. Also, users unaffected by this change will still be able to see other users' replies and could then share screenshots of said replies which would defeat the point of this exercise - by using language boundaries means it's less likely that such a thing would happen (e.g. if this was Sweden and Norway instead of US+Japan, would you really bother to go through Swedish tweets to find replies for other users?)

In short: it sounds like they're uncertain about pushing this change through on everyone, and want to trial it first.

Twitter has gone-back on ideas they've thrown around before - like when they said they would hide other users' tweets' Like +RT counts.


Actually makes it easier to see outside the echo chamber of a thread


I'm sure this feature isn't going to be abused


It will be used to hide legitimate criticisms of political tweets mainly. This is a win for capital protection, nothing more.


wonder if they'll add a feature to autoblock someone who has hidden a reply of yours


Bot time.


This feature in many ways reminds me of shadowvoting on HN


Will politicians be able to use this feature?


I would imagine so. As long as you don't block people (so they can't see your stuff when logged in, or reply under that handle), I don't think there are any potential legal issues. They can say anything they want, and anyone else can see anything anyone said (if they click through to see it). So it's a minor inconvenience, but I don't think this presents the same issue as blocking others.


Much needed to detoxicate Twitter itself.


So will a court have to rule that Trump cannot use this?


[flagged]


Yes.


Yes


Ye


Y



seY


I usually make it a point to retweet my trolls, a technique due to Andrew Breitbart.

Another great technique is to ask simple, intelligent questions.

My trolls seem to lack stamina.

ADDED:

The other extreme thing I like to do on social media when people are trying to troll me is offer sincere forgiveness.


[flagged]


Have you ever considered that classifying people who disagree with you as a race of subhuman monsters is "amplifying the negativity you find in this world"?

If a bored fourteen year old really offers a rhetorical threat to your argument, get new arguments.


The word "troll" in this context originates from the fishing technique, not the monster.


Proof?



How can someone be a verb? I'm a eating. Doesn't make sense.


"Troll" is both a verb and a noun. Originally it was just a verb: "to provoke responses by metaphorically dragging bait", but it soon broadened to mean "somebody who provokes responses by metaphorically dragging bait". And in modern times the meaning has broadened still further to include direct insults, which is a needless loss of precision in language because there was already the perfectly good verb "flame" for that meaning . But even after all the extension of meaning, the origin of the word is still the dragging of fishing bait.




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: