Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Twitter- optimised for abuse

weekwoman.wordpress.com

Post reply on HN
Print view View original

11–20 of 52 posts

Re: Twitter- optimised for abuse

#11

Post by icebraining »

chroma wrote: post #8

Many of those tweets contain direct threats of violence. They need to be reported to law enforcement, not flagged for Twitter to deal with. Also, it seems unusual to hold Twitter responsible for their users' speech. UI tweaks may reduce the amount of hatred seen, but they are not going to solve this problem. Lastly, I spent some time perusing other posts on this blog. They are not furthering any discussion, to put it…

Many of those tweets contain direct threats of violence. They need to be reported to law enforcement, not flagged for Twitter to deal with.

They're not mutually exclusive. Twitter should still take all reasonable steps to prevent abuse.

I don't think HN has much to contribute.

I completely agree; I cringe every time something vaguely related to feminism or gender equality is posted here. I'm not sure if I should flag it or not, though.

Re: Twitter- optimised for abuse

#12

Post by mtrimpe »

chroma wrote: post #8

Many of those tweets contain direct threats of violence. They need to be reported to law enforcement, not flagged for Twitter to deal with. Also, it seems unusual to hold Twitter responsible for their users' speech. UI tweaks may reduce the amount of hatred seen, but they are not going to solve this problem. Lastly, I spent some time perusing other posts on this blog. They are not furthering any discussion, to put it…

Why could we not contribute? At it's core this is about the needs of a very small but important group of users not being addressed.

It only took me a few minutes to think of various things we could do to aid them ranging from a UI more optimized for blocking and reporting; services to offer pre-filtering of comments.

The one concrete thing that Twitter could do (and should do in my opinion) is helping actual criminal charges to be filed for hate speech (inciting to violence.)

I'm still convinced that the only reason this persists is because the perpetrators believe, and so far they're right, that they will not be meaningfully punished for behaviour like this.

Re: Twitter- optimised for abuse

#13

Post by stormbrew »

danso wrote: post #5

This is one (sad) illustration of how Twitter is so unique a communication service, because of its particular access-control details. The fact that it's so hard to block/stop abusers is a consequence of the way Twitter leaves communications "open" and loose...with Facebook, you can't stalk/directly-abuse someone without them giving you some access (or revoking Facebook's default access). With email, you can target an…

One thing that you can't do without being logged in on Twitter is retweet, which has apparently been a powerful tool in coordinating these account bombings. This is why the block feature being changed to ignore had such a backlash -- not just because it turned the name into a lie, but because blocking was a somewhat useful tool for preventing the amplification of these attacks.

Blocking and ignoring are different, and both useful.

Re: Twitter- optimised for abuse

#15

Post by clearly »

icebraining wrote: post #11
chroma wrote: post #8

Many of those tweets contain direct threats of violence. They need to be reported to law enforcement, not flagged for Twitter to deal with. Also, it seems unusual to hold Twitter responsible for their users' speech. UI tweaks may reduce the amount of hatred seen, but they are not going to solve this problem. Lastly, I spent some time perusing other posts on this blog. They are not furthering any discussion, to put it…

Many of those tweets contain direct threats of violence. They need to be reported to law enforcement, not flagged for Twitter to deal with. They're not mutually exclusive. Twitter should still take all reasonable steps to prevent abuse. I don't think HN has much to contribute. I completely agree; I cringe every time something vaguely related to feminism or gender equality is posted here. I'm not sure if I should flag…

I'm genuinely interested to hear you say that, why do you cringe? Because it doesn't affect you and you don't perceive there to be a problem?

This is clearly a technology story, and I'm interested to hear more from anyone at Twitter how their release process works in this regard. I saw Kent Beck talk at facebook a while back and he spoke about how when working on a site like FB, you literally do have peoples lives in your hands. Releasing that software is a very hard problem. The example he gave was if they accidentally allowed public access to someones private mail- it could literally lead to deaths.

Re: Twitter- optimised for abuse

#16

Post by mtrimpe »

jval wrote: post #7

Would it be that difficult to add in an 'Abusive Off' feature? You can just flick it off and any tweets containing blacklisted words vanish. Would be extremely valuable for people who come under attack, I imagine. A lot of people who are obscenity averse might like to enable it too. Heck, I might even enable it myself when reading the activity on some hashtags.

Or allow users to give your reaction on tweets and have a simple machine learning algorithm block, filter and hide tweets for you based on that.

The real issue here though is that these trolls know that there will be no legal consequences for their actions. That needs to change for anything to change and Twitter does have a key role to play in that... they're just understandably reluctant to do so.

Given that there are real costs attached to addressing the issue (not just legal but also in terms of engagement statistics) as a rational business they will probably delay addressing the issue until the costs of ignoring it (in terms of bad publicity) exceed the benefits.

Re: Twitter- optimised for abuse

#17

Post by tomp »

clearly wrote: post #15
icebraining wrote: post #11

Earlier quoted context omitted.

Many of those tweets contain direct threats of violence. They need to be reported to law enforcement, not flagged for Twitter to deal with. They're not mutually exclusive. Twitter should still take all reasonable steps to prevent abuse. I don't think HN has much to contribute. I completely agree; I cringe every time something vaguely related to feminism or gender equality is posted here. I'm not sure if I should flag…

I'm genuinely interested to hear you say that, why do you cringe? Because it doesn't affect you and you don't perceive there to be a problem? This is clearly a technology story, and I'm interested to hear more from anyone at Twitter how their release process works in this regard. I saw Kent Beck talk at facebook a while back and he spoke about how when working on a site like FB, you literally do have peoples lives in…

Because such stories are usually knee-jerk reactions, badly argumented, badly researched, un-objective, and assume some opinions that not everyone agrees with are facts. Furthermore, pointing this out can inspire rage and insults from a part of the online population. Finally, they induce a flame-war with no clear victors and no apparent benefits for the victors. They don't really encourage discussion, just as most static-dynamic do not.

Re: Twitter- optimised for abuse

#18

Post by jacquesm »

It's like shouting in a crowded theater. Only this theater is about the size of the globe and the audience is incapable of distinguishing between a joke and a real threat so everything is looked at in the worst possible way.

Internet communications are hard, tone-of-voice is meta information that we jokingly try to add with and tags but when those are omitted real trouble can ensue.

Anonymous threats are rarely worth the value of the medium used for their conveyance.

Re: Twitter- optimised for abuse

#19

Post by pjc50 »

mtrimpe wrote: post #12
chroma wrote: post #8

Many of those tweets contain direct threats of violence. They need to be reported to law enforcement, not flagged for Twitter to deal with. Also, it seems unusual to hold Twitter responsible for their users' speech. UI tweaks may reduce the amount of hatred seen, but they are not going to solve this problem. Lastly, I spent some time perusing other posts on this blog. They are not furthering any discussion, to put it…

Why could we not contribute? At it's core this is about the needs of a very small but important group of users not being addressed. It only took me a few minutes to think of various things we could do to aid them ranging from a UI more optimized for blocking and reporting; services to offer pre-filtering of comments. The one concrete thing that Twitter could do (and should do in my opinion) is helping actual criminal…

It's not even a small group of users, there's a lot of abuse about and for women it often takes a very nasty tone.

There is occasionally legal action taken about tweets, but it's so rare and inconsistently applied that it manages to be a threat to free speech but not to actual abuse.

Re: Twitter- optimised for abuse

#20

Post by gaadd33 »

icebraining wrote: post #11
chroma wrote: post #8

Many of those tweets contain direct threats of violence. They need to be reported to law enforcement, not flagged for Twitter to deal with. Also, it seems unusual to hold Twitter responsible for their users' speech. UI tweaks may reduce the amount of hatred seen, but they are not going to solve this problem. Lastly, I spent some time perusing other posts on this blog. They are not furthering any discussion, to put it…

Many of those tweets contain direct threats of violence. They need to be reported to law enforcement, not flagged for Twitter to deal with. They're not mutually exclusive. Twitter should still take all reasonable steps to prevent abuse. I don't think HN has much to contribute. I completely agree; I cringe every time something vaguely related to feminism or gender equality is posted here. I'm not sure if I should flag…

How can Twitter prevent people from posting abusive messages? Should they have some sort of keyword filter that prohibits specific words and regexs that match addresses?
Post reply on HN