[content warning: violence, trauma, harassment]
The irony is not lost on me that I incited a tidal wave of harassment from Reddit’s cesspit of toxicity because I went in there to talk about working on anti-harassment software. It was the perfect case in point for why we’re building what we’re building at Block Party. And all the same, it was a torturous way to re-learn a lesson I’ve already had beaten into my psyche from more than 15 years of online bullying, hate, abuse, and stalking.
I was so foolishly optimistic about how a Reddit AMA might go
When someone first suggested the idea of going on Reddit, I instinctively recoiled from the idea, knowing the site’s reputation — but I ignored my intuition and last week I went in naively optimistic about doing an AMA (“ask me anything”) question and answer session in the subreddit r/IAmA/. I thought it would be a good opportunity to share more about how I came to be working on Block Party and to engage in genuine discourse about the problem of online harassment and our product thinking on how to solve it. Perhaps I would get questions about what I’ve learned from building and running a distributed team, or other lessons from working in Silicon Valley and then leaving for less tech-saturated locales. Or about diversity and inclusion, especially as I just released a course for startup founders. When I announced on Twitter that I would be doing a Reddit AMA, I even got a few good questions there: about COVID and re-opening, collaboration with social media platforms, crowdsourcing of Block Party’s filters.
The site tagline is “Reddit gives you the best of the internet in one place”. I shudder to think what the worst of it is, Reddit is vile beyond belief
I barely started the AMA on Reddit before the troll brigade arrived. They quickly overtook the page with bad faith “questions”, gotchas, twisted accusations, and insulting and abusive comments. They coordinated to downvote all of my answers into oblivion, as Reddit’s platform hides content that has been excessively downvoted. As I struggled to surface my answers, they taunted me about my alleged silence: “She has no answer because her views are wrong.” “She didn’t answer because she hates asian americans..” It felt like a violent mob had ambushed me and was forcing me to the ground, muzzling me, and violating me. Typically authors will start their AMA, let questions come in for a while, then write answers for an hour, maybe longer if they’re generous. I tried, I really tried, even as I write this now I’m crying as the feeling of trauma resurfaces, for four hours I tried, fighting through the abuse to write candid, thoughtful answers about what we’re working on with Block Party and how we’re trying to solve online harassment, I addressed willful misinterpretations of our privacy policy, questions about filter bubbles and polarization, data or lack thereof on the problem of online harassment, the role of machine learning in content moderation, diversity and inclusion in hiring, diversity and inclusion outside of America, activism, how Asian Americans figure in racial inequity conversations — no matter how hard I tried, it didn’t matter. Even for innocuous content, like the list of books I’m currently reading, I got downvoted into negative territory and an asshat reply saying I wasn’t actually reading those books and I cherry-picked those to be over-the-top. The harassment spilled over into Reddit direct chats, onto Twitter, into waitlist signups for Block Party where people submitted flagrantly racist and other abusive content.
Even worse than the harassment on my AMA itself was Reddit’s infuriating response: blaming me and refusing to take responsibility
Worse than the trauma of the AMA itself, which landed on the Reddit homepage and got up to 4k+ comments, the overwhelming majority of which were toxic and vile, was the response. My god, the response. From Reddit, I got only blame, gaslighting, a refusal to take any responsibility, and a bland declaration that they do not condone harassment on their site. Then they put the onus on me to report any “policy-breaking comments” so they could investigate — yes, because of course it should be the person being targeted with abuse who’s responsible for wading through the vileness to clean it up. One r/IAmA/ moderator went onto Twitter as well to blame me for apparently doing it “wrong”. There were more who told me it wasn’t Reddit’s fault that people just didn’t like my answers. Friends privately reached out to Reddit employees, including some that I know and (used to) regard as friends. A number of these people work in policy and product, areas directly responsible for a situation like this. The responses ranged from a reiteration of putting the blame on me, to silence and refusal to acknowledge the situation, much less express any sympathy. The absolute abdication of accountability has been the most angering part of the whole experience.
None of this is new. Reddit’s problems with harassment have been documented widely for many years. I am no expert on Reddit as I previously steered far clear of it and apart from my mistake last week to do an AMA, will do so in the future; but it is clear that they do not care to make even the most basic fixes to check the metastasis of the hate on their platform.
The sleight of hand Reddit is attempting in order to cast the blame on me is this: An error in how my AMA was posted prevented the pathetically meager anti-harassment protections they do have from kicking in (though all those would have done was prevent my answers from being downvoted into oblivion, it would not have been any defense against the thousands of trolls showing up to post harassing content). The claim is that I didn’t know how to use the site so it’s my fault. To be very clear, though, the error occurred on the platform side. And even if I had made a mistake because Reddit’s usability is so poor and it is a fundamentally user-unfriendly product, ultimately that is still on them. If Redditors pick up on this post and use technical nit-picking to try to discredit me, that only further proves my point. All of this finger-pointing from Reddit company, the official Twitter account, the employees, the moderators, the users, is merely a means of distracting from their own culpability.
Try to follow, if you want. Here’s the “technical issue”: Someone else can schedule an AMA on your behalf, but if so, then it may post from their account, such that the OP does not match your account even if yours is marked correctly as the person conducting the AMA. In this scenario, you have no downvote brigade protections, and even as the AMA author, your answers can be disappeared. You’re shit outta luck, sorry.
I find Reddit’s deflection of the blame enraging. First, my teammate and I did nothing wrong in how we scheduled and conducted the AMA. We filled out all the boxes correctly. She put her username in the box for who was scheduling the AMA, and my username in the box for who was conducting the AMA. If you need any proof of that, my account was flaired correctly as the one doing the AMA. Second, it was never our responsibility to reverse-engineer Reddit’s pathetic anti-harassment protections to understand that there is only one way to avoid the silencing that comes from coordinated downvoting. Third, the true problem is that Reddit’s voting mechanisms have this pernicious silencing effect on people from minority or marginalized communities. This is well known to mods as well as researchers. One sympathetic mod did let me know that this happens elsewhere too: For example, Black scientists in r/science/ getting so downvoted by racists that their posts disappear, even when they were expressly invited to do panel discussions on their areas of expertise. This admission of a pervasive issue leads to the next point: Fourth, it’s not actually that fucking hard to fix this. I am a software engineer who was literally one of the first engineers at Quora, another consumer web scale user-generated content site with answers, comments, voting, and ranking. Guess what, all you need to do to protect an answer from being collapsed is to add a not-collapsible boolean to it. Yes/no. That’s it. If moderators see an answer being inappropriately hidden, someone being unfairly silenced, it is possible, and not just possible but infuriatingly simple, to fix it. (Yes, I know that this is only one small tool that is needed and there are much broader issues, but even so, it seems stupid obvious to me to build this.) I cannot believe the condescension of the multiple responses I got from Reddit team and mods that the situation was “not possible to fix”, “it’s a technical issue”. Fifth, even if somehow the most basic engineering change is impossible to prioritize, help docs and other copy are easy to change. At very least, the scheduling tool should have an aggressive, big, bold warning to AMA participants that the system has this vulnerability.
Online harassment is a form of violence
Online harassment is a form of violence. What I experienced on Reddit was a coordinated, sustained act of violence. And not only did the authorities and responsible parties not stop it, they gaslit and blamed the victim. My own friends, the ones who could do something, merely looked away. They were unable to offer even an acknowledgement of the situation, much less a modicum of sympathy. I am still shaking in anger and disappointment.
If you have never experienced violence in this way, it can be hard to understand how traumatic it is. Nowhere is safe. I know what it’s like to be physically stalked and threatened, to be fearful of my security in the offline world, and that is bad enough. Sometimes I wake up in a panic wondering if I forgot to lock my door, and I still start when I hear unexpected sounds outside my flat. But at least there are places that I can lock myself in and hide away, and I can be with friends who protect me. Online, the same platforms where I get encouragement and solidarity from supporters are the ones where I receive harassment as well. Do you know the feeling when someone makes a remark that bothers you and it sticks, even when you try to shrug it off? It might not even be directed at you, or maybe it is but you know it’s not true, just a throwaway insult, but it keeps running through your head. Imagine that, multiplied by thousands, a mob feeding off its own energy, shouting you down with perverted, nasty things, choking your protests into silence, declaring their freedom of speech. Your supporters might be trying to get through, too, but it’s hard to hear them through the onslaught of abuse. The psychological toll is terrible. I have dealt with online harassment for a long time, but this past week crushed me. I haven’t been able to sleep through the night, despite the Ambien, anti-anxiety medication, CBD, melatonin, essential oils, meditation, anything I could possibly try. My resting heart rate spiked 20% the day of the Reddit AMA and my physiological state remains extremely poor.
I also have to live with the fear and possibility that the online harassment will escalate, that I will have yet more reason to fear for my physical safety. Just in the last few days there have been multiple news stories about people being shot dead, their murderers groomed and goaded on by toxic communities like Reddit’s, a noxious trail of online hate in their wake. People who have the power, influence, and ability to do something need to do it, not perform allyship on social media while disclaiming responsibility for harm directly attributable to their action or, more likely, inaction. I can only hope that they act faster in the future when danger and death is imminent.
Welcome to the life of being a startup founder
I have still not been able to take a break. I am the solo founder of a very early stage startup and things are always go-go-go-go-go.
The latest urgent, p1 TODO for Block Party is one triggered entirely by the Reddit situation. If you hazard a glimpse at the AMA page, there are strenuous objections to our privacy policy, and although it was a standard one that we picked up from our original legal counsel and we do absolutely none of the nefarious things that people accuse us of, they twisted reality so much I felt we should address it. Numerous Redditors even pointed out that I used to work at Facebook as proof that I must be violating user privacy and selling user data — the cruel joke is that I was so put off by what I saw during my Facebook internship that I once wrote an answer on Quora decrying the cavalier disregard for people’s privacy, but my answer got downvoted into the negative by Facebook employees and collapsed. For the past week, I’ve been spending late nights with our current lawyer to draft a new custom privacy policy that makes it clear exactly what we do. For a startup like Block Party, which is at such an early stage and still has so few users, it feels like a waste of investor money, not to mention my time, to be combing over these legal documents, but I want to be clear on our commitment to putting the user first.
If we must grasp at silver linings, I am only more certain of the importance of what we’re working on at Block Party
The Redditors in my AMA perfectly illustrated the very need for “troll management” software, and though we didn’t need it, we did get additional validation for many of Block Party’s product intuitions and directions.
My biggest takeaway is that it’s quite clear platforms will not solve the problem of online harassment. The Reddit company stance has been that it’s not a priority to implement the product or engineering fixes that could turn back the tide on the toxic sludge. This doesn’t leave the moderators with much to work with, even when some good ones do care to try. In the case of my AMA, they were conspicuously absent, and though they very belatedly took some action to lock my AMA thread and do minor cleanup, all it did was leave the thread frozen with pages and pages of bad faith argumentation and misinterpretation and wrongful accusations that you had to scroll through before you could find anything I’d written.
To a further point about content moderation, there is a big difference between what’s not so egregious to be in violation of policy and what makes for reasonable, good faith discussion and a healthy community. Some of the most perverse trolling comes from people who are just clever enough to twist words around in a pseudo-intellectual fashion, with a pretense at being “genuine”, good at toeing the line and knowing how to avoid being removed or banned. As a participant in a community, or simply a guest, I have no interest in engaging with those people. I am interested in civil discourse that brings hard questions, contrary viewpoints, and challenges to my assumptions and thinking, but just because a platform has disallowed the shouting of racist and misogynistic slurs and will sometimes belatedly chastise people for it does not mean it has created a productive space for the interchange of ideas and information.
As for the most toxic, terrible comments on my AMA that were eventually removed, their absence from the thread now does nothing to remedy the damage that has already been wreaked, and the destruction of the evidence is only a setup for more gaslighting. I’ve already been deluged with the hate. I’ve already suffered for it, my mental well-being poisoned. Perhaps after I publish this essay, it will spur the moderators to remove even more of the bad comments. But it will only be worse that everything is gone. Abusers and harassers know this tactic well: Attack, then destroy the evidence. The most wretched of them will come back around to torment further. Even curious bystanders coming by later will not understand the violence and trauma that has happened, instead choosing to doubt the victim, wondering if things were really that bad. I know I’m supposed to document everything, but I cannot sacrifice any more of my mental health to do so.
And we have yet another data point for how pathological it is that platforms put the burden of managing abuse on those targeted by it. Reddit not only makes it so others cannot help you, but if they do try to—as in the case of my teammate helping me to schedule the AMA — you are subject to even worse terrors than if you went at it alone. At least for catching the harassment spillover to Twitter, I had Block Party to filter my @mentions, and my teammate was able to go through everything that collected in my Lockout Folder to block, unmute, or keep muted as appropriate. It was a relief to still be able to use Twitter without letting the trolls keep tearing down my mental health. And though some of those harassers fell back on their usual trick of deleting their Tweets, Block Party is keeping the receipts for me.
A deep thank you to those who gave me support, acknowledgment, and validation through the most traumatic episode of my life
I am grateful to the many people who reached out to offer their support and encouragement, the people who posted indignation and outrage on my behalf, the people who threw me “upvote parties” to try to rescue some of my answers, all these people who cared. I cried, in a good way, to read the messages from people who told me what my work has meant to them, how they have directly benefited from it or been inspired to pursue their careers. I cried, too, when reading the messages from people acknowledging and validating what happened. I needed to hear that I’m not crazy, not overreacting, that none of this is okay, that none of this is my fault. I haven’t been able to respond to every message but to all of you who have sent positivity my way, thank you.
Unfortunately, I have a creeping sense of dread that things are only going to get worse as the brokenness of the Internet becomes more and more overwhelming and the toxicity of the tech industry forces out those who have the heart and integrity to try to make it better. As for myself, I am doing all I can to stay in the fight. I’m doing everything I can to try to make things better, not just for me but for everyone who must endure this almost unbearable heaviness of online existence.
