“A Honeypot for Assholes”: Inside Twitter’s 10-Year Failure to Stop Harassment
buzzfeed.com
buzzfeed.com
From a technical perspective, letting essentially anyone say anything is easier to handle. The engineering resources that would otherwise go towards advanced content filtering and detection, etc can now just focus on making the platform scale. That works out well when a company is growing like crazy in the early days.
From a business perspective, the hate-spewing users are "really engaged with the product." Having a lot of users in this category is a great thing to tell investors. Also, the growth from these users is something investors like, too. So it's also a win from a business perspective.
I'm guessing that Twitter's leadership made a decision early on. Something along the lines of "let's go with the free speech thing for now. It lends itself to efficient allocation of engineering resources and improving key metrics." When you're growing as fast as Twitter in the early days, it's not necessarily a bad choice. However, the current growth stagnation and other negative issues surrounding the company is really starting to expose the downsides of that haphazard decision.
You take on risks when you start announcing your private thoughts to the widest possible public audience. Those risks go along with being any kind of public creator. Other than because no one would really be interested in my tweets anyway, this is precisely the reason why myself and others exercise our free choice not to tweet. We don't want a trail of statements for which we are publically held to account, even if by harassers, racists and misogynists, if we can't - and shouldn't - control the audience for those tweets. If you want a limited, curated audience, go somewhere else.
It's the expectations we've projected on twitter. A forum for anonymous messaging run by people who believe in free speech is NOT an appropriate place for an online presence as a celebrity or business. You really shouldn't tie it to your real name. It's perfectly fine for there to exist 4chan/twitter-esque cesspools of hate and wretched abuse at every turn. It's not perfectly fine for people to, as part of their job, log into thousands of racist hate-spewing garbage messages targeted at them every day, or have their online identity plastered with it.
This is the kind of place that twitter is. Some are sheltered from the storm and don't notice it. Partly because you aren't the target audience for random acts of hate (maybe you're a white male in tech), and mostly because you just haven't gotten unlucky enough to attract the attention of certain hordes of hate spewing monsters. But if you use twitter to network, you're diving into the cesspool. If you invite others to network with you on twitter, you're inviting them to do the same. If you post a twitter account as part of your online presence, you're broadcasting to the world "yup, I'm part of this".
And that's a real problem for twitter. You can't have both unrestrained free speech and civilized communication.
I mean the actual harmful harassment. The article mentioned that one of the co-stars of the new Ghostbusters quit online media due to the cesspool, but there are many many more instances of people taking the endless microphone of online social media and making it horrible. Though it is incredibly rare, some people are in fact driven to suicide due to the targeted harassment.
Now we can debate facts and the amount of thickness your skin should be until the Zuck finally kicks the bucket. But the point I want to make is that sometimes the cesspool gets away from itself and steps over some invisible line into true harassment that can't be defended to a lot of people as simple freedom of speech (lawyers assemble!). I love that we have freedom of speech in this country, and I know that it is not meant for the things that you or I like to hear and that it is explicitly for the lowlifes and assholes to say what they want to. Still, I don't know where that line is, but I do know that it gets crossed a lot; and that from the media I consume, the line tends to be crossed a lot more if you are female or black.
So, I'll defer to the people that have been bitten by the cesspool, as I have not been in their shoes and have had to stare at their phones while hate comes blasting in and all they are trying to do is order a pizza. If a lot of the people that have had these experiences say that Twitter et al need to be better, then I think I have to trust them.
You can have conversations without it becoming a racist, bigoted mess.
You might think that the downvote mechanism here helps somewhat, and it does, but on any large community with voting such as Reddit the ability to downvote turns the community into an echo chamber where only the largest group gets a say and any dissenting voice is vanished. I dare say we see the same thing here on HN.
I imagine if there was competition and a "Twitter but with anti-trolling mechanisms" network sprung up most people, and the vast majority of women and ethnic minorities would flock to that network.
I think the point is that these risks are unevenly distributed. When I announce my thoughts on Twitter my risk is lower by virtue of the fact that I'm a dude, I'm straight, or any number of other qualifiers. It's not that no one experiences any risks when announcing themselves in public, it's that the risk is dramatically and severely unevenly distributed.
The unequal distribution of risk (and consequently, the unequal distribution of vitriol) is IMO the entire point of the article, and has two main takeaways:
- This results in some demographics experiencing relatively little abuse, and therefore assume a proportionally larger share of the conversation as those who experience high abuse quit.
- The system, by virtue of the fact that it's built mostly by people who experience low-risk when using it, is in a large way blind to high-risk members of its population, and will not develop the necessary tools to combat abuse (see also: Reddit).
> "If you want a limited, curated audience, go somewhere else."
This is very, very evidently happening. Twitter's user growth has stalled, its total user base has plateaued, despite competing social networks continuing to grow rapidly. There are, of course, more than enough anecdotes of both high- and low-profile users fleeing the platform due to abuse, and the aggregate numbers reflect this.
It's also worth noting that the younger generation has overwhelmingly chosen social networks with either more limited audiences, or more abuse management tools (see: Snapchat, Instagram).
The market is speaking, though it makes me wonder if some middle ground was possible - a "public announcement" social network where the most extreme ends of the abuse spectrum could be contained. Twitter fulfills a valuable niche, and there is a certain amount of sadness that it's really not working out.
I think the main difference is they tend not to start campaigns to systematically 'sanitise' Twitter nor claim to speak on behalf of their entire race/gender. They deal with it in different, much quieter ways.
For example, it somehow manages to ignore the many harassment campaigns against various white people (e.g. Justine Sacco, Pax Dickinson, Brandon Eich, Milo) who's politics don't conform (or based on a single tweet, appear not to conform) to what is acceptable.
Or it brings up Leslie Jones, but then completely ignores all her racist tweets and calls for follower to harass others: http://www.breitbart.com/tech/2016/07/20/double-standards-le...
That sure is weird.
[edit: clarified my point better.]
Who are you trying to kid? More and more, your comments on HN are becoming a parody of themselves. You have real arguments to make. I don't like most of them, but at least I can respect them. This, though, is more horseshit.
That's a fairly extreme claim, and I'd hate to be in a position to evaluate its accuracy.
> @whitebecky1776 bitch I want to tell you about your self but I'm gonna let everybody else do it I'm gonna retweet your hate!! Get her!!
And the Twitter rules say:
https://support.twitter.com/articles/18311
> Harassment: You may not incite or engage in the targeted abuse or harassment of others. Some of the factors that we may consider when evaluating abusive behavior include:
> ...
> if the reported account is inciting others to harass another account;
> ...
Just FYI.
What's up with all the concern trolling? Is everything ok?
Dismissing them as "racist trolls" is a bit odd. Milo is a well known journalist, who doesn't seem much worse in his way than any other muck-racking person going after views.
Some minor google-fu yields this:
http://www.themarysue.com/milo-yiannopoulos-permanently-bann...
In which they claim that un-named people have engaged in racist baiting of Leslie Jones. And that those people are believed (for reasons un-named) to be Milo's followers. And that Milo is on record as not denouncing those people. And therefore Milo should be banned from Twitter for being a racist troll.
If Milo violated Twitters terms of service, then by all means show which inappropriate tweet(s) Milo made.
Otherwise, does no one really see a problem with such a tenuous chain of logic behind banning him?
Leslie Jones calling out 'white people' is not racism. Black Americans expressing negative sentiments about white Americans is definitionally different from racism. I wouldn't say it's necessarily the most productive way to challenge white supremacy, but it's not of a piece with the centuries old and still strong system of dehumanizing, oppressing and killing people of a specific race in America.
No, its not.
> I wouldn't say it's necessarily tasteful on Jones' part, but it's not of a piece with the centuries old and still strong system of dehumanizing, oppressing and killing people of a specific race in America.
That doesn't mean that its "definitionally different from racism", that just means its not an aspect of a particular course of racism.
Ideally, we want to call out and stop racism before it manifests in a course like that, rather than only afterwards. (Which is not to say that a course with that history, and the kind of continuation that anti-black racism in the US still has, is not a more significant concern -- just that racism doesn't become racism only after it has successfully managed to justify generations of crimes against humanity.)
Yes, it is. This is the crux of what most white Americans don't get about race in America. Anti-black racism originated from a thesis of racial inferiority. The anti-white "racism" the Breitbart article cites clearly doesn't have a founding thesis that "white people are inherently inferior". It's actually just a critique of the myriad ways white 'culture' perpetuates anti-black (and anti anything but straight white male) sentiments and beliefs. Certainly, I don't think pithy tweets are the best way to advance these critiques, but they're not being written for the benefit of cerebral tech bros like us, they're being written to offer some relief and solidarity for others whose lives are materially affected by white bullshit.
No, its not. (Isn't this fun?)
> This is the crux of what most white Americans don't get about race in America.
Having spent much of my childhood being called a "nigger" by White Americans (and some, though less, being called a "zebra" or "oreo" by Black Americans), I really don't need any lectures from anyone about what "most White Americans" don't get about race in America.
And while there are certainly many things that many (or even most) White Americans don't get about race in America, what some (but, IME and AFAICT more generally, not most) Black Americans (and, even more paternalistic white-knighting White Americans) don't get is that negative racial stereotyping both is and promotes racism no matter the race of the proponent of the stereotype or the race targeted by the stereotype.
EDIT: I wanted to add a comment on this:
> Certainly, I don't think pithy tweets are the best way to advance these critiques, but they're not being written for the benefit of cerebral tech bros like us, they're being written to offer some relief and solidarity for others whose lives are materially affected by white bullshit.
The fact that a racist utterance provides some relief to an audience that has had negative experiences doesn't make it any less racist. Heck, one of the main uses of anti-black racism by white economic and political elites in the US has been to offer some relief and (illusion of) solidarity to poorer white Americans suffering from adverse conditions by directing the blame at blacks in general.
You're right, racial stereotyping of any kind is bad; I was wrong to argue otherwise. I got on my paternalistic white-knighting horse because I made the (racist?) assumption that you were white, and I'm inclined to argue against other white people accusing others of racism. I guess it comes from my own toxic white guilt. But is it a completely wrongheaded impulse, to try to police other white people who are trying to argue that American racism goes equally both ways? That's what the Breitbart article is implying, that Jones was being just as toxic as Milo, isn't it? My argument that anti-white racism is definitionally impossible may be overly radical, but don't you have to somewhere take account of the fact that anti-white racism doesn't have nearly the same material impact on it's targets and isn't really a structural part of political and economic systems in America?
It is not wrongheaded to attempt to counter arguments that the harms from being targeted by racism [0] in America are equally distributed among racial groups, I would agree.
> My argument that anti-white racism is definitionally impossible may be overly radical, but don't you have to somewhere take account of the fact that anti-white racism doesn't have nearly the same material impact on it's targets and isn't really a structural part of political and economic systems in America?
Sure, but I think that there is both a problem with any concept of "racism" that goes so far as to say it is not racism (as I discuss more in my response to your post discussing your use of such a definition) and, more tactically, that simply calling anti-white racism "not racism" does more to obscure, rather than effectively addresses, why and how there are significant differences, in the present circumstances that exist in the US, in the impacts of anti-white vs. anti-black racism (which I think you identify quite well.)
[0] I was going to say "harms from racism", but given that there are many ways in which (among other non-blacks) poor whites are harmed by anti-black racism manipulated by elites, the harms from racism are probably at least more equally distributed than the direct harms from being targeted by racism are, and the latter is, really, I think the issue of concern here.
I.e., twitter isn't against harassment, they are just against Milo and vague hints of harassment are a convenient excuse.
1. "Prejudice on the basis of race or color." Reference: pretty much any dictionary.
2. "Oppression based on racial hierarchy." Harder to find a reference, but here's one: http://www.diversityinc.com/ask-the-white-guy/does-playing-t...
Note that under definition #1 black people can be racist against white people. Under definition #2 that's not possible, because black people are (or at the very least historically have been) oppressed.
I have seen #2 used to make categorical statements like "black people can't be racist." Which may not be an Orwellian dystopia as @adekok said in a sibling comment, but it is inflammatory to people using definition #1.
The problem with this definition is that it conflates racism -- which is problematic in any case -- with an external condition which interacts with racism and makes it more damaging (and, worse, it treats that external condition -- which in reality is a continuously-valued property -- as binary.)
By doing so it obstructs discussion of the continuum of problems associated with racism and serves to normalize some racism (which may be less harmful racism in the conditions in which it originates, but as those conditions change for what would otherwise be the better, becomes more harmful without the actual content of the racism changing.)
It also has the additional unfortunate side effect that people more distantly following the theorists adopting such a radical definition pick up the conclusion "Black people can't be racist" and lose the context of the understanding of the conditions justifying, in the particular definition the theorists are applying, why negative racial stereotyping of whites by blacks isn't racism, and just taking it as a categorical privilege of blacks (regardless of their position and privilege in a particular milieu) to engage in racial stereotyping and discrimination, because "black people can't be racist".
The dictionary definition of racism would be the one most people agree with. Which then means that black people making anti-white statements are being racist.
The alternative is an Orwellian dystopia in which the meaning of words depends on who says them.
> ... they're being written to offer some relief and solidarity for others whose lives are materially affected by white bullshit.
Which is itself a racist comment according to the dictionary definition. Engaging in racism in order to stop racism means that you're no better than the racists you oppose.
I'm firmly of the belief that the ends don't justify the means. The means help determine the end.
To that end, we all should oppose racism. All racism. And not just oppose racism by people we don't like, and support racism by people we like.
If there is true information that is being reported only by certain sources, it is perhaps worth thinking about why. See this article by Freddy Deboer on that topic:
http://fredrikdeboer.com/2016/07/20/whos-in-and-whos-out-tha...
"People who gleefully trashed Justine Sacco complain about pile-ons; people who say doxing is wrong get others fired from their real-life jobs. There are no principles; there’s only who you’re cool with and who you aren’t."
Please provide proof.
They've already made their money back.
https://mic.com/articles/113226/forget-your-team-your-online...
who objectively decides what is and isn't?
This is circular reasoning.
If I can define X as abusive (where X is something I don't like), then I am a person getting abused. So now that allows me to define whatever I want as abusive?
And while I get your point about circularity, it pretty much has to be that way. For example, if I threatened to stab a random person with a syringe, that would be abusive. If I threatened to stab my brother with a syringe, that's an inside joke from when we used to play "Battlefield Vietnam" together.
Which is pretty much what the article was getting at when they talked about context mattering more than content.
Pointing out that "abuse is whatever an abused person defines it as" is an argument built on quicksand is very much not bikeshedding.
Ok, in that case lets also catch the low hanging fruit of black men raping white women. And if you say "a black man asking out a girl who doesn't like him isn't rape" then you are guilty of bikeshedding. <- Spot the fallacy yet?
If circularity is valid then as a victim I define names that start with "brei" to be harassment. You can tell I'm a victim because someone with a name starting with "brei" replied to me, which makes this valid. <- Spot the fallacy yet?
Consider the possibility that if you can't prove your point without deliberately resorting to logical fallacies, that your point might be wrong.
But since you seem to think that there is a fully-cohesive, non-circular, objective way to judge what is harassment or abuse, please enlighten me as to what it is.
If you fail to do so, then somebody will probably rape and kill your entire family. <-- Not abusive, because there isn't a peer-reviewed Theory of Abuse available.
edit: Just for the hard-of-sarcasm, the last bit is to illustrate the ridiculousness of the argument, not an actual threat.
There is a vast amount of abuse of Twitter to be dealt with long before we get into any territory where subjectivity is required.
It's that whole people behind an anonymous screen are assholes because they can thing.
Humanity has always lived on subsistence and never flown, until it fixed those things.
> Twitter as a platform doesn't do anything differently than the old IIRCs or forums when it comes to this.
The vast majority of forums I've used had responsive moderators and the ability to remove trolls (including correlating concurrent or sequential sockpuppets and reapplying sanctions on new accounts). IRC tends to be more spartan, but even there you could require authentication, autokick abusers and ban IPs.
Of course both were intended as communities (of sorts) not trying to sell their growth, so removing a bunch of assholes wasn't considered an issue by anyone.
So Instagram clearly does something differently, and from the perspective of at least one person, better.