Muting words is dumb. It's not really complicated to write n1gger instead of nigger.
Suspending accounts is the dumbest, though—the account of an abuser is likely to be in the block/mute list of hundreds of people. By suspending his account, the user will just create another and continue annoying the very same people that had blocked him already.
I wonder who in Twitter, Inc. uses Twitter at all.
I'm already on some block lists. I'll run into some journalists who have me blocked. Never interacted with them. I hardly interact with anyone, but some block lists just take the entire list of followers from people and add them.
these exist, but because 'annoying account' is a subjective decision, while 'advertisement url' is an objective decision, it is difficult to create a list with a clear brightline between harassment and disagreement
Sure, but there are libraries to identify visual similar words. And yes, you would fight them with being even more obscure, but at some point they'd almost become unreadable.
How about a bayesian approach, like email?
Contains masked content +3
New account +4
No profile pic +5
Follows accounts with low reputation +4
Greater than out imaginary threshold of 10 so maybe hide the mention from whoever's timeline?
There are degrees of belief attached to each aspect and the proposal works exactly like existing acknowledged-as-bayesian approaches eg spamassassin does. So how isn't it?
The inclusion of example values was to act as an example. Nothing was stated about where they were derived from. It would be logical to derive them from actual frequency in real life moderation scenarios. Please don't make bad faith assumptions.
> Sorry, I didn't mean to presume they weren't based on values extracted from real data. I'm worried about people misusing CS terms, but I can see I kind of jumped the gun there.
An obnoxious user can create another account but it takes work and the new account won't have the same reach as the old one. Of course there are strategies to circumvent this, but the economic costs for the troll steadily increase. Once the costs exceed the propaganda benefit, it drains resources from the propaganda effort.
I think it works pretty well. Granted, Metafilter is also heavily moderated to prevent conversation derails and general abuse, but that's also what the money pays for (not like money is a problem for TWTR, or is it?)
I don't believe the payment is used towards enforcing identity (MeFi values privacy and anonymity when needed) and locating abusers, but if things got really bad I don't see why that would be a huge problem.
I'm glad to see that Twitter is responding to what users want and need, and acknowledging that there's no easy answers here.
Any sort of blocking solution is hard, I think akin to Facebook's "fake news" problem. There's no easy way to determine what news is "fake" or not. But giving people tools that they can use to manage their own space in their own way is an important step forward.
I hope that this will encourage more people to speak out on twitter about injustice and abuse, since it gives them some ability to shield themselves from any backlash.
Facebook has un-follow and see less from tony. Twitter is just launching mute, which is similar to un-follow. I'd like both services to offer "don't show me links from tony" because that's really what I want when I choose "see less from."
I've been clicking "hide all from xyz" on article shares for a few years now and almost have all the most popular link sources covered. A blanket setting to mute article shares from some or all of my friends would be most helpful. But something tells me that these article shares are a big part of FB's engagement strategy.
One type of abuse that seems oddly popular here in Mexico is via a trending topic. I suspect that some kind of mafia group with a lot of followers are testing their weapons (since creating a trending topic is a very good business) by insulting women, minorities and even disabled people.
Blocking, muting and reporting users isn't the answer when you have right in your face a hateful hashtag. You could look for the one who started the trending topic and report it, but with thousands of mentions, you'll have to scroll a lot.
I actually thought this "progress" was about reporting trending topics, or finding the one who started it. And then, once again, we see twitter turning the blind eye.
I don't even know how to see the trending topics while mobile. To say that a list of topics on a web version of twitter only is worse than targeted abuse from users seems silly to me. I think it should be trivial to block the trending topics list from loading in your browser, even.
Oh, there it is. When I tap the search icon, I'm tapping again on the input box immediately to bring up my saved searches, so I'm not focusing on that trending list (or seeing it most times) because I'm waiting for my saved searches to appear.
It's a start. There will be ways around the mute filters by using alternative spellings, substituting in similar unicode characters, etc.
There needs to be a "quality" score for a Twitter user and an option to mute tweets from "low quality" users. Quality could be determined by utilizing an algorithm similar to Google Page Rank. Retweets, likes, and follows by "quality" users would be considered in a user's quality rank. Reports of hateful conduct would be weighted in as well.
It's too bad that it has taken this long for Twitter to even -start- to address the abuse. They are lucky they have no direct competitors.
My take on this is that, besides Twitter's general product stasis, the reason they've been conflicted on this is that the experience for new users and power users is completely different. For a new user, any response is great and Twitter wants to encourage that. For a power user, they have more than enough responses ("randos in my mentions").
I do feel that greater control over "notifications" is the obvious approach here and glad they're introducing controls like 'mute conversation'. Medium has an interesting approach with comments where you can see all comments vs. author-approved comments. Twitter probably doesn't want to make their UI complex with a lot of options but more granular control on replies (maybe even an ability to disable replies and quote tweets) would probably be useful.
Unsubscribing from notifications for a specific conversation is the only useful feature here. Muting words/hashtags from showing up in your TL would have been useful. Regarding "online abuse", shared blocklists were much more useful than locking or suspending accounts, that's not going to help at all.
Also I wonder what exactly falls under 'behavior that incites fear about a protected group' and 'repeated and/or or non-consensual slurs, epithets, racist and sexist tropes, or other content that degrades someone'. Will Twitter enforce politically correct tweets? Is NSFW content going to be banned because it offends some people?
I think it's reasonable to ban calls for violence against individuals, but I don't think banning all sorts of hateful content is a good idea and it's not going to help removing it anyway. If people don't want to see certain content, by all means, let them hide it from their feed and notifications.
As long as Twitter applies a double standard to abuse against white men, it'll be a useless political echo chamber. Allowing "#killallmen" posters to get away with their drivel while bringing the banhammer on anyone challenging the SJW narrative is itself taking a strong and IMHO obnoxious political stance. Twitter is unbearable
All you need to do is look at the people twitter invite to their conferences, and associate with at their executive level to know that Twitter will only get worse in these regards.
Twitter is an anonymous comment section on steroids. How about some kind of real user verification beyond email addresses? (i.e. Phone verification) It could be an optional step, but then provide the ability to ignore any Twitter content that doesn't come from these more-verified accounts?
As stated previously, there's no easy answer. Even when validating a telephone number, then everyone would use Twilio to create multiple "validated" accounts.
Using Twilio to get around the verification may be obvious if you are a HN reader, but most people have no idea what Twilio is. Phone verification is certainly not a bulletproof solution, but I have to feel it would help a lot.
And it doesn't have to stop at phone, either. You could multiple levels of verification, like Airbnb. Then as a user you could further filter how strict you want your tweet exposure to be.
Yep. Phone verification certainly isn't bulletproof. But reducing the deluge of abusive posts by troll accounts featuring Pepe memes as their profile image even 10 or 20% would be tremendous, in my view.
Every hurdle helps. Like OpenBSD `pledge`. It doesn't stop everything and processes like bash can fork/exec to unwrap the pledge wrapper. But it helps get in the way of return oriented programming.
Theo de Raadt had a good description of why every hurdle helps, but I can't find it. :(
This is actually a really good idea, as it would ensure that there is a citizen behind each Twitter account. Currently this lack of control hinders implementation of more effective systems for combating online abuse such as a social credit score system.
I find the easiest way to addressing online abuse, is not using social media, and installing comment section blockers for browsers. I get the content I want, and I don't have to be distracted by the nonsense.
I'd wager that you are probably not the target of a lot of this abuse, though - so you likely wouldn't have seen much of it anyway, and wouldn't be as affected by it. (For instance, I'm betting @imadefood doesn't get a lot of tweets from eggs calling you any four-letter profanities that begin with "cu".
You've chosen to opt out. That works for you, which is great, but you should be aware that it isn't addressing the problem of online abuse in any way whatsoever.
Online abuse isn't simply seeing 'the nonsense'; online abuse is when people send hateful messages to you in response to what you post. Suggesting people shouldn't post if they don't want to see the abuse you're cutting off their voice, which is not an acceptable solution. People should be able to say things online if they want to. They shouldn't have to see the horrible responses. That's not to say people shouldn't be able to post horrible responses (although there's a strong argument that they shouldn't), just that there needs to be tools to filter what you receive. That's what Twitter is building. It's a good thing.
Yeah all that is great until you realize the people who control the filters are biased and, for example, allow a hashtag like #killallwhites to exist. Where's the diversity councils and the racism SWAT teams for that?
"We’re enabling you to mute keywords, phrases, and even entire conversations you don’t want to see notifications about"
So, instead of fostering more exchange of ideas and information, we're just going to have a bigger echo chamber to help perpetuate groupthink, misinformation and convenient lies?
I'm not trying to be rude but does no one else see why this is a problem?
If you end up in the crosshairs of a user with many aggressive/abusive followers, seeing their tweets in your mentions ruins the SNR of your twitter experience.
There's a reason why celebrities have had the "quality filter" available for ages. A very high volume of tweets makes it more difficult to use twitter to communicate.
People don't want to see the world as it is, they want to see the world as they see it.
Twitter has decided that instead of focusing on showing the world as it develops, instead of closing the digital divide and creating a space for the "new media"...
They want to be a social network. And make money from it. (yeah, right)
I want a better Twitter to exist someday. I might make one myself (and no, GNU Social isn't good enough yet). But that Twitter would have to be focused around things that are happening, around global occurrences and real news. Not some stupid internet catfights that could be solved by turning one's phone off for a day.
I don't really use twitter so I can't picture some of these interactions, but how can people be getting abusive notifications?
To get abusive notifications, doesn't the unfortunate recipient have to follow the abuser first?
Does Twitter really allow anyone to @mention anyone else without the recipients having to confirm it? Or does it in fact notify the recipient with the full text of the comment, so harassers are exploiting the approval mechanism?
I'm asking because this sounds like a recipe for disaster, which is frankly what they've been having.
If we accept the notion that Twitter resembles a public bulletin board, then unpleasant (and potentially, hate) speech can and will come at anyone, and just like in real life, often the recipient has no methods of recourse other than trying to disregard it -- not making a value judgment whether this is fair, but this is how it actually happens. Blatant threats to life and property can be reported to the authorities, but most real-life verbal harassment cannot be pre-emptively filtered.
On the other hand, if we respect that each user's timeline, notification space, and knowledge graph is controlled by the user, unsolicited, unapproved @mentions are disruptive by nature. But there doesn't seem to be a way in Twitter right now to disable unsolicited @mentions or make them require the mentionee's approval.
So Twitter, by its behavior, seems to take the first stance. Why, then, does any Twitter user have an expectation that abuse targeting them can be limited by the company?
Twitter really brings out the worst in people. While accounts aren't ID verified, this sort of abuse won't be affected. Let's just face it, people really suck behind an anonymity curtain.
I think the key is that some accounts are verified and some people choose to use their real names. Bullying seems lower for the average Facebook user because everyone can see exactly who's calling you a fat loser. And bullying on 4chan and the like seems pretty low because everyone's anonymous so no one knows if you're actually fat or not. Twitter is this weird overlap where some groups share who they really are, but the harassers are free to hide behind anonymity.
> A Blizzard employee going by the handle Bashiok revealed his real name, Micah Whipple, on the forums in the course of the debate. WoW Riot responded by quickly finding and publishing Whipple's address, phone number, age, the names of his family, his Facebook page and lists of his favourite music and movies, to illustrate the privacy and identity theft issues raised by Blizzard's decision.
No thank you. Make your own social network and have it be fully verified if you'd like, but I wouldn't consider joining it for a second.
1) All anti-bias statements should include body-weight / shape. Far too many people are abused/discriminated against on the basis of their weight.
Referring to this sentence:
> Our hateful conduct policy prohibits specific conduct that targets people on the basis of race, ethnicity, national origin, sexual orientation, gender, gender identity, religious affiliation, age, disability, or disease.
2. What steps are you taking to address the seeming divide between left/right abuse enforcement? For instance, AssassinateTrump was trending for a time. Will you offer evidence showing that you are banning abusive political speech from the left and right?
They are right to focus on notifications. That's where the annoying or hostile crap accumulates (abuse, spam).
I think it would be valuable to split notifications into two areas. Low and high value notifications. High-value would tend to be stuff associated with accounts you follow, and maybe unfollowed, verified accounts judged as 'high quality'. Everything else - interactions from 'randos' - gets filed as low-value.
Nothing needs to get hidden from users who want to see every time anyone interacts with them.
To elaborate: only high value notifications activate OS notifications (i.e. your phone beeping.)
Low-value notifications simply accumulate in a separate chronological list. There would be a visual signification in the Twitter UI when you have new unseen stuff there.
This will make spam accounts that 'like' random tweets, to get people's attention, less effective.
The downside is it raises a barrier to possibly valuable new connections being made between strangers. But not a hard barrier. It'll take some extra effort for a message-recipient to engage - they need to proactively check their low-value-notifications page.
Twitter would prefer a solution that preserves a high engagement level. So I don't think they'll adopt my great idea.
Should the Twitter app on my device serve me, or Twitter Inc.? This seems like a conflict of interest!
Imagining a superior, more harmonious social networking ecosystem: separation between developers of client interfaces and message-delivering infrastructure. This is the approach Urbit is trying...
Twitter's free, ad-supported model, is not financially successful. It may not survive, whether it's disrupted by an alien computing ecosystem or not. In the course of evolving the microblogging concept, other desirable aspects may die too, like Twitter's monopoly status (and thus its near-universal reach). Maybe we'll end up with hundreds of separate networks, with varying user policies and performance qualities.
It might seem small, but it's a big step in the right direction. Every block/report on a user/tweets could be rolled up and adjust a spam/abuse score for that user. That would allow twitter users to set a threshold for users that show up in their feed (tweets blocked for that user, akin to google's safe search).
One challenge of this is attempts by groups to harm a user's score by blocking/reporting non-hate/spam. This requires a block or report to be weighted by the blocker/reporter's credibility. It's definitely doable, * Warning Shameless plug* a product I've worked on https://www.inversoft.com/products/profanity-filter handles these situations well. Without some type of credibility tracking it's usually incorrect to blindly trust reports/blocks.
No matter what system is put in place, their are large, determined, and even some well funded groups which will game any and all systems to modify, censor, or punish others if you allow them the power to 'group moderate', especially in the political space - which is a very large part of twitter.
Just the combination of the headline and the image below it are enough to get me to close the tab on this one. I'm getting really tired of the constant accusations of racism, and this immediately set off my 'upcoming bullshit' detector.
86 comments
[ 4.3 ms ] story [ 159 ms ] threadSuspending accounts is the dumbest, though—the account of an abuser is likely to be in the block/mute list of hundreds of people. By suspending his account, the user will just create another and continue annoying the very same people that had blocked him already.
I wonder who in Twitter, Inc. uses Twitter at all.
How about a bayesian approach, like email?
Contains masked content +3
New account +4
No profile pic +5
Follows accounts with low reputation +4
Greater than out imaginary threshold of 10 so maybe hide the mention from whoever's timeline?
There are other machine learning approaches that would also learn from data - specifically, a Bayesian approach would use Bayes theorem.
Nothing about my post ever suggested the value was arbitrary. You've just assumed that.
And in any case, a Bayesian approach would require more than a handful of criteria.
No problem, happens to the best of us.
I'm not going to apologize for reading the words you wrote, nor for not being able to read your mind.
Why can't Twitter do what Metafilter does? $5 to sign up and a 24 hour delay on any posting.
How is it working out for Metafilter?
I don't believe the payment is used towards enforcing identity (MeFi values privacy and anonymity when needed) and locating abusers, but if things got really bad I don't see why that would be a huge problem.
Any sort of blocking solution is hard, I think akin to Facebook's "fake news" problem. There's no easy way to determine what news is "fake" or not. But giving people tools that they can use to manage their own space in their own way is an important step forward.
I hope that this will encourage more people to speak out on twitter about injustice and abuse, since it gives them some ability to shield themselves from any backlash.
> for a few years now
Each time I see that option, I think how crazy I could get blacklisting every FB under the sun. It sounds like you've tried!
Blocking, muting and reporting users isn't the answer when you have right in your face a hateful hashtag. You could look for the one who started the trending topic and report it, but with thousands of mentions, you'll have to scroll a lot.
I actually thought this "progress" was about reporting trending topics, or finding the one who started it. And then, once again, we see twitter turning the blind eye.
There needs to be a "quality" score for a Twitter user and an option to mute tweets from "low quality" users. Quality could be determined by utilizing an algorithm similar to Google Page Rank. Retweets, likes, and follows by "quality" users would be considered in a user's quality rank. Reports of hateful conduct would be weighted in as well.
It's too bad that it has taken this long for Twitter to even -start- to address the abuse. They are lucky they have no direct competitors.
I do feel that greater control over "notifications" is the obvious approach here and glad they're introducing controls like 'mute conversation'. Medium has an interesting approach with comments where you can see all comments vs. author-approved comments. Twitter probably doesn't want to make their UI complex with a lot of options but more granular control on replies (maybe even an ability to disable replies and quote tweets) would probably be useful.
There was a Black Mirror episode about this in season 3. Seems it was on point.
Also I wonder what exactly falls under 'behavior that incites fear about a protected group' and 'repeated and/or or non-consensual slurs, epithets, racist and sexist tropes, or other content that degrades someone'. Will Twitter enforce politically correct tweets? Is NSFW content going to be banned because it offends some people?
I think it's reasonable to ban calls for violence against individuals, but I don't think banning all sorts of hateful content is a good idea and it's not going to help removing it anyway. If people don't want to see certain content, by all means, let them hide it from their feed and notifications.
And it doesn't have to stop at phone, either. You could multiple levels of verification, like Airbnb. Then as a user you could further filter how strict you want your tweet exposure to be.
4chan isn't a bunch of computer illiterate folks, either.
Theo de Raadt had a good description of why every hurdle helps, but I can't find it. :(
It's like fighting neighborhood crime by drawing your curtains.
Ignoring/hiding online abuse, is very effective at shutting down the culprits. If no one is effected, they can't do anything.
Online abuse isn't simply seeing 'the nonsense'; online abuse is when people send hateful messages to you in response to what you post. Suggesting people shouldn't post if they don't want to see the abuse you're cutting off their voice, which is not an acceptable solution. People should be able to say things online if they want to. They shouldn't have to see the horrible responses. That's not to say people shouldn't be able to post horrible responses (although there's a strong argument that they shouldn't), just that there needs to be tools to filter what you receive. That's what Twitter is building. It's a good thing.
So, instead of fostering more exchange of ideas and information, we're just going to have a bigger echo chamber to help perpetuate groupthink, misinformation and convenient lies?
I'm not trying to be rude but does no one else see why this is a problem?
There's a reason why celebrities have had the "quality filter" available for ages. A very high volume of tweets makes it more difficult to use twitter to communicate.
Twitter has decided that instead of focusing on showing the world as it develops, instead of closing the digital divide and creating a space for the "new media"...
They want to be a social network. And make money from it. (yeah, right)
I want a better Twitter to exist someday. I might make one myself (and no, GNU Social isn't good enough yet). But that Twitter would have to be focused around things that are happening, around global occurrences and real news. Not some stupid internet catfights that could be solved by turning one's phone off for a day.
To get abusive notifications, doesn't the unfortunate recipient have to follow the abuser first?
Does Twitter really allow anyone to @mention anyone else without the recipients having to confirm it? Or does it in fact notify the recipient with the full text of the comment, so harassers are exploiting the approval mechanism?
If we accept the notion that Twitter resembles a public bulletin board, then unpleasant (and potentially, hate) speech can and will come at anyone, and just like in real life, often the recipient has no methods of recourse other than trying to disregard it -- not making a value judgment whether this is fair, but this is how it actually happens. Blatant threats to life and property can be reported to the authorities, but most real-life verbal harassment cannot be pre-emptively filtered.
On the other hand, if we respect that each user's timeline, notification space, and knowledge graph is controlled by the user, unsolicited, unapproved @mentions are disruptive by nature. But there doesn't seem to be a way in Twitter right now to disable unsolicited @mentions or make them require the mentionee's approval.
So Twitter, by its behavior, seems to take the first stance. Why, then, does any Twitter user have an expectation that abuse targeting them can be limited by the company?
It's great for when you want to talk to celebrities. Unfortunately, oftentimes celebrities don't want to talk to you.
Remember Blizzard Real ID? http://www.eurogamer.net/articles/blizzard-scraps-real-id-fo...
> A Blizzard employee going by the handle Bashiok revealed his real name, Micah Whipple, on the forums in the course of the debate. WoW Riot responded by quickly finding and publishing Whipple's address, phone number, age, the names of his family, his Facebook page and lists of his favourite music and movies, to illustrate the privacy and identity theft issues raised by Blizzard's decision.
No thank you. Make your own social network and have it be fully verified if you'd like, but I wouldn't consider joining it for a second.
Referring to this sentence: > Our hateful conduct policy prohibits specific conduct that targets people on the basis of race, ethnicity, national origin, sexual orientation, gender, gender identity, religious affiliation, age, disability, or disease.
2. What steps are you taking to address the seeming divide between left/right abuse enforcement? For instance, AssassinateTrump was trending for a time. Will you offer evidence showing that you are banning abusive political speech from the left and right?
I think it would be valuable to split notifications into two areas. Low and high value notifications. High-value would tend to be stuff associated with accounts you follow, and maybe unfollowed, verified accounts judged as 'high quality'. Everything else - interactions from 'randos' - gets filed as low-value.
Nothing needs to get hidden from users who want to see every time anyone interacts with them.
Low-value notifications simply accumulate in a separate chronological list. There would be a visual signification in the Twitter UI when you have new unseen stuff there.
This will make spam accounts that 'like' random tweets, to get people's attention, less effective.
The downside is it raises a barrier to possibly valuable new connections being made between strangers. But not a hard barrier. It'll take some extra effort for a message-recipient to engage - they need to proactively check their low-value-notifications page.
Twitter would prefer a solution that preserves a high engagement level. So I don't think they'll adopt my great idea.
Should the Twitter app on my device serve me, or Twitter Inc.? This seems like a conflict of interest!
Imagining a superior, more harmonious social networking ecosystem: separation between developers of client interfaces and message-delivering infrastructure. This is the approach Urbit is trying...
Twitter's free, ad-supported model, is not financially successful. It may not survive, whether it's disrupted by an alien computing ecosystem or not. In the course of evolving the microblogging concept, other desirable aspects may die too, like Twitter's monopoly status (and thus its near-universal reach). Maybe we'll end up with hundreds of separate networks, with varying user policies and performance qualities.
One challenge of this is attempts by groups to harm a user's score by blocking/reporting non-hate/spam. This requires a block or report to be weighted by the blocker/reporter's credibility. It's definitely doable, * Warning Shameless plug* a product I've worked on https://www.inversoft.com/products/profanity-filter handles these situations well. Without some type of credibility tracking it's usually incorrect to blindly trust reports/blocks.