8 ms·
Twitter’s Policy Reboot: The Good, the Bad, and the Ugly
- afarrell 11y agoI've seen multiple people claim that it would be easy for Twitter to ban harassment on Twitter.
- mahouse 11y agoSame here, what I haven't seen is how would that be done.
- geofft 11y agoHave you read Sarah Jeong's The Internet of Garbage? http://www.amazon.com/Internet-Garbage-Sarah-Jeong-ebook/dp/B011JAV030/ http://www.amazon.com/Internet-Garbage-Sarah-Jeong-ebook/dp/... https://play.google.com/store/books/details/Sarah_Jeong_The_Internet_Of_Garbage?id=i24nCgAAQBAJ https://play.google.com/store/books/details/Sarah_Jeong_The_... (DRM-free)
- hobs 11y agoThe same way it would be easy for crypto companies to build a backdoor, in that if you dont understand the problem, it is easy to solve. edit: (and remain secure)
- afarrell 11y agoMore that it is easy if you don't care about throwing away other things you value in pursuit of that goal
- jaydigital 11y agoQuick question to people that know more about these things than I do (pretty much everyone on HN): Is there a way that Twitter can prevent users from creating another Twitter account to continue to send threats and horrible messages? The worst messages we see on Twitter always seem to be from new accounts with 3 followers, clearly throwaway accounts. Ban one, they start another.
- bcoates 11y agoNot without making it much harder to onboard new users. User growth (Monthly active users) is a business goal Twitter reports to investors so that's a non-starter.
- mahouse 11y agoThey already force people in some ip ranges (for example tor exits) to do mobile # auth.
- criddell 11y agoI can't think of a reasonable way of blocking people with bad intentions from creating Twitter accounts. But there are other things they could do. One approach would be to expand their verified user program to people like me (ie the proletariat) and then have an option to switch the blacklist model of blocking people to a whitelist model of showing only tweets from verified users. Or they could make a whitelist based on people I follow or that friends follow (or friends-of-friends etc...). Or maybe just get serious with analytics and sentiment analysis and don't show me stuff that is likely a troll.
- ascagnel_ 11y agoFor a service that's trying to grow, that would force them to put all new users into an "unverified" bucket, which will probably be a turn-off to those users.
- criddell 11y ago> that would force them to put all new users into an "unverified" bucket But that's exactly how it is now. I personally like the verified user option but I can understand how some wouldn't for reasons like the one you give. Frankly, I think Twitter's focus on growth is a big part of what got it into this problem in the first place. I'd rather see them focus on quality than growth and I think expanding the verified user program does that. The measure-of-connectedness option is a little gentler and would probably fix a pretty big piece of the problem.
- Grue3 11y agoThe new Twitter policy is already used to mass-silence activists by dedicated (paid) teams of trolls. http://euromaidanpress.com/2016/01/09/twitters-new-policy-misused-by-pro-kremlin-accounts-to-attack-top-ukrainian-bloggers-saveuatwi/ http://euromaidanpress.com/2016/01/09/twitters-new-policy-mi...
- WillAbides 11y agoI'm curious to know how this is happening. Perhaps the people who handle complaints in Russian and Ukranian are Putin supporters and bring their personal animosities to work. It would be hard for Twitter to hold accountable judgement calls in languages management doesn't understand.
- wcummings 11y ago>I'm curious to know how this is happening. Seems pretty straightforward if you believe the claim that the abuse complaints are coordinated, even if they're "legitimate". Both pro-Ukraine and pro-Soviet users are violating the new rules, but people are systematically seeking out and reporting violations on pro-Ukraine twitter accounts and the same isn't happening to pro-Soviet accounts.
- jqm 11y agoMaybe because there aren't actually any Soviets anymore?
- Bossman 11y agoYeah. There have been a few instances of this already. Not just trolls, but journalists, people discussing politics, etc. I feel like Twitter trying to bend over to appease the people who want to combat unpleasantness online will have more unintended consequences.
- binarymax 11y agoI don't envy Twitter's position right now. They are stuck between (1) trying to renew their platform to monetize and grow, (2) facing huge amounts of criticism on how they dont properly handle abuse, and (3) expected to keep their reputation of openness still left over from the 'Arab spring'. It is doubtful that they can survive without sacrificing at least one.
- danielcarvalho 11y agoShould be "threats of violence" not "violent threats". Since people think anything can be deemed violent nowadays.
- oldmanjay 11y agoI don't get nostalgic for much, but one thing I miss about the earlier internet was everyone's ability to dismiss internet tough-assery as ridiculous, overblown stupidity. I guess that's the difference between a pioneer town and a safe space on a university campus.
- krstck 11y agoI'm particularly sad about the evolution of the word "troll" to mean "anything someone else doesn't like". The idea of "flamebait" has also completely disappeared, and we've lost the ability to roll our eyes at it.
- s73v3r 11y agoMost people didn't have to deal with it on a daily basis, and hardly anyone had threats that were deemed credible enough to actually happen. Sorry, but it's not all ridiculous, overblown stupidity anymore. Some of these people seem stupid enough to actually follow through on their threats.
- oldmanjay 11y ago"Seem" is an interesting word here, as well as the phrase "deemed credible." Both indicate that my point is not being rebutted. I'm not denying that the scale of the problem has grown to the point that rare events (someone following through on Internet bluster, in this case) will occur. I'm simply remembering a simpler time, when people didn't clamor for the sanitization of their daily Internet lives.
- acdha 11y agoNot wanting to receive threats of violence against you or family members is hardly clamoring for sanitization. Sure, most of those threats aren't actually going to happen but a real human has the stress of having to evaluate each one and hoping they didn't get it wrong. It's also not as if the only two possibilities are “actual stalker attacks you” and “nothing happens”. SWATing, spamming friends/family/coworker with hateful propaganda or photoshopped porn, placing embarrassing ads with the victim's actual contact info, etc. are all things that happen in the real world and don't require some blowhard to leave their house or risk persecution anywhere near as much as a serious physical attack would entail.
- protomyth 11y agoThe most interesting reveal to me has been the the "verified" flag isn't really a verification at all.
- golemotron 11y agoEspecially after the removal of verification from noted cultural theorist Milo Yiannopoulos.
- makomk 11y agoYeah. I don't particularly like the guy, but it's not like he stopped being Milo Yiannopoulos (sadly).
- _cpancake 11y agoTwitter's rules say they can remove your verified status at any time for violating their ToS. It's pretty obvious that Milo has violated their ToS many, many times, so he's not verified anymore. Don't know what the big deal about that is.
- sib 11y agoI would say that the big deal is, if that is the policy, then verification does not represent "verification" so much as "you play by the rules"... As has been said elsewhere, just because Milo violated some policies, he didn't stop being Milo.
- talmand 11y agoI play by the rules and I haven't been verified.
- protomyth 11y ago"Don't know what the big deal about that is." Verified was originally sold as twitter saying the account belonged to a verified person or business. We could trust it was that person or organization and not some fake account. This isn't true anymore and they should rename it to something else. The other part is that their are apparently some advertising requirements associated with verified for businesses.
- xyzzy4 11y agoI think the only feasible way Twitter could stop abuse is if it charged $100 per signup, and then ban people on the slightest whim. I don't see how anything else would work. Of course the downsides of a plan like this are so large that it wouldn't ever be done.
- Kristine1975 11y agoThe forum Something Awful does something like this: They charge $10 per signup. When you get banned you have to pay again to get back in.
- scholia 11y agoEven $1 would remove a lot of harassment because a lot of people would be less abusive if thought it would be easier to find them and hold them legally responsible for their comments.
- geofft 11y agoI expect there are plenty of abusive people on Twitter who could afford that $100 several times. There's no particular reason to believe that being abusive on Twitter is correlated with being poor.
- mahouse 11y agoI looked up a picture of her and I don't think I want to read it.
- geofft 11y agoI am confused by this statement.
- Nadya 11y agoHopefully nobody shoots the messenger (or in this case, translator): She has dyed hair which is increasingly becoming a sign of narcissistic Tumblr types. The hair dye is seen as "pay ME attention, I'm a special snowflake!" This is because the group of people often referred to as SJW's consist of a large majority of women with dyed hair (reds, pinks, blues). In other words: it has become a stereotype. Therefore, the other user is making a presumption that due to her dyed hair coupled with the title of her book she is an oversensitive Tumblr-type who isn't worth the time to read. I tried to keep this explanation as neutral as possible, hope it clears up your confusion.
- mahouse 11y agoAnd also she's written for Vice, The Verge, Forbes, The Guardian, Slate, and WIRED (according to her website). Translation: I expect a very very biased book I'm not interesting in reading.
- Nadya 11y agoWell you mentioned her picture, not her history. :) I was just looping them in on why you might say that. I can see where you're coming from about the extreme bias. Only two missing from that list are Salon and Jezebel, yeesh.
- mahouse 11y agoI admit I saw the blog list just when I posted it, it was enough for me to look at the pic to know what to expect ;P
- gottam 11y agoI don't understand about the scope of this problem, maybe someone can tell me but is it really that big of a deal? What horrible things are being typed out that can't possibly be ignored or dealt with in other ways?
- s73v3r 11y agoIf you're constantly receiving a barrage of horrible, hateful, violent threats, then it becomes really hard to ignore. Some people may have thicker skin than others, but really, no one should have to deal with that. And quite frankly, I see no reason why anyone should be entitled to say those kinds of things on someone else's platform.
- gottam 11y agoI find that hard to believe. Trolls who make threats on the internet (and not simply having differing opinions) typically make up a very small minority so they should be manageable by a small ban list (and if they aren't - what kind of stuff are you spewing out that makes so many people waste their time getting angry with you?) It sounds more to me like even after getting banned, the minority of trolls can just make new accounts and resume their harassment, so the solution here isn't human intervention but a technical solution to disincentivize trolls from making new accounts.
- DanBC 11y ago> and if they aren't - what kind of stuff are you spewing out that makes so many people waste their time getting angry with you? The Bank of England phased out their £5 notes. English bank notes have portraits of famous people on them. The old note had a woman, the new note had a man. That meant that all English bank notes had portraits of men; there were no women on the notes. Caroline Criado Perez the Bank of England to reconsider this decision, and asked for a woman to appear on at least one of the notes. For that she was bombarded with messages. One man was sending about 50 an hour, for about 12 hours, across multiple accounts. Other people sent fewer tweets, but those tweets were far more menacing. Here's what the judge said about 2 people who sent threats of rape and murder: https://www.judiciary.gov.uk/wp-content/uploads/JCO/Documents/Judgments/r-v-nimmo-and-sorley.pdf https://www.judiciary.gov.uk/wp-content/uploads/JCO/Document... > Here there were a series of communications (not just a single one as referred to in the guideline) where extreme language was used and substantial fear was caused. > Indeed it is hard to imagine more extreme threats (death, rape, and worse [Sorley]) and rape [Nimmo]. The harm is, and (despite what is said on your behalf) must have been intended to be, very high. > The fact that they were anonymous heightened the fear. The victims had no way of knowing how dangerous the people making the threats were, whether they had just come out of prison, or how to recognise and avoid them if they came across them in public. And about blocking: > The messages were posted on a number of different sites, and continued after being blocked and or warned
- CM30 11y agoThe biggest issue with the policy (and one which is not really mentioned here) is enforcement. Doesn't really matter what the policies are, if Twitter is very, very loose when it comes to actually enforcing them. For example, I've seen a few people get banned or suspended because they said something that made a popular user (or in some cases, one a Twitter member of staff agrees with) uncomfortable, whereas someone who supports said staff member's political views or has more followers can get away with murder. For example, there have been people who said something loosely associated with GamerGate who got suspended until they removed the tweet, whereas certain journalists and media figures could try and falsely accuse their enemies of serious crimes and get in no trouble whatsoever. It was also a certain Twitter employee who got Milo unverified for ridiculous reasons. Twitter's policy reboot is already worrying, but what it really needs to do is be even handed with its punishments towards those that break the rule. It shouldn't matter if you're a nobody or a celebrity with millions of followers; break the rules, get suspended or banned. It shouldn't matter if your political views align with Twitter staff or not. Same deal. Twitter needs to examine what it's policy on freedom of speech and stuff is, but then also enforce it equally, however unpopular that may be.
- exstudent2 11y agoTwitter has also started censoring their search suggestions down the same political lines you mention. You no longer get auto-suggest for #gamergate or any Milo related topics. Ironically they're also blocking #twittercensorship. You can see for yourself here: https://twitter.com/search-home https://twitter.com/search-home As far as I know this change wasn't announced (seems to have rolled out over the weekend in response to the Milo incident). It does not bode well for freedom of speech on Twitter. Edit: to clarify the results are still there if you manually do the search, but the hashtags are no longer in the search suggestions.