The Burden Of Everyone Hating Your Website
Over the weekend, the calls for digital infrastructure company Cloudflare to block Kiwi Farms, a notorious message board run by Joshua “Null” Moon, finally hit the point of no return. As I write this on Monday morning, Cloudflare has blocked the site and Kiwi Farms has been effectively removed from the internet, with its last remaining members marooned on a Telegram channel. The drive to get Cloudflare to block Kiwi Farms was led by Clara “keffals” Sorrenti, a trans Twitch streamer from Canada. It is not hyperbolic to say that Sorrenti, aided by the numerous activists and journalists that amplified her, has done what was once thought impossible.
Kiwi Farms was started when Moon was an 8chan admin in 2013. It was originally created as a board to discuss the CWCki, a wiki created by 4chan and 8chan users to doxx and obsess over Christine Weston Chandler, the web comic artist known as Chris-Chan (that’s what the “CWC” stands for). Its original name, CWCki Forums, was shifted to Kiwi Farms, and the method of obsessive documentation of the internet’s various low-level (and not so low-level) internet personalities was then expanded beyond just Chandler to any “main character of the internet” the community noticed. If you did earn yourself a thread on Kiwi Farms there was no way to remove it. The site also became increasingly focused on doxxing trans creators over the years.
Explaining what Kiwi Farms, as a website, has actually done is tricky, but it largely operated the same way right-wing influencers like Libs Of TikTok currently do, by pointing an internet firehose at individuals. Kiwi Farms users would make a thread about someone and then other users would stalk, SWAT, and threaten those individuals. Moon’s defense has always been that those things did not happen on the site itself, but elsewhere online. Cloudflare’s Matthew Prince wrote a post about the decision to block Kiwi Farms over the weekend, citing an “immediate threat to human life” as the company’s main reason for finally pulling the plug.
“Kiwi Farms itself will most likely find other infrastructure that allows them to come back online, as the Daily Stormer and 8chan did themselves after we terminated them,” Prince writes. “And, even if they don't, the individuals that used the site to increasingly terrorize will feel even more isolated and attacked and may lash out further.”
In another blog post from last week, Prince said that Cloudflare’s security services, many of which are free and are used by an estimated 20 percent of the entire internet, should be thought of as a utility. “Just as the telephone company doesn't terminate your line if you say awful, racist, bigoted things, we have concluded in consultation with politicians, policy makers, and experts that turning off security services because we think what you publish is despicable is the wrong policy,” Prince wrote.
Which is a good line. I’m sure people who are old enough to remember when telephones weren’t computers love it. But I’m not really sure it works here. Telephones are not publishing platforms, nor are they searchable public records. Comparing a message board that has around nine million visitors a month to someone saying something racist on the telephone is, actually, nuts.
But, more broadly, I don’t even think this is a free speech issue. Cloudflare isn’t a government entity and it’s not putting Kiwi Farms members in jail. In fact, it seems like some users have done that themselves. A German woman seems to have accidentally exposed her real identity amid the constant migration of the site and now may be charged for cyberstalking. Instead, Cloudflare, a private company, has removed their protection from the site, which allows activists and hackers to DDoS it, taking it down.
In Prince’s blog post announcing Cloudflare has blocked the site, he writes that neo-Nazi blog the Daily Stormer and 4chan-spinoff and QAnon hub 8chan were both able to come back online in some form after Cloudflare blocked them. This is true, but to say that either site is bigger or even the same as they once were is laughable. The Daily Stormer is basically only accessible now via a TOR browser and 8chan has now shattered into a million spinoffs, unable to consolidate users like it once could. Which is really what we’re talking about.
Websites are not similar to telephones. They are not even similar to books or magazines. They are street corners, they are billboards, they are parks, they are shopping malls, they are spaces where people congregate. Just because you cannot see the (hopefully) tens of thousands of other people reading this blog post right now doesn’t mean they’re not there. And that is doubly true for a user-generated content platform. And regardless of the right to free speech and the right to assemble guaranteed in America, if the crowd you bring together in a physical space starts to threaten people, even if they’re doing it in the periphery of your audience, the private security company you hired as crowd control no longer has to support you. To me, it’s honestly just that simple.
The big question now is about what will happen next. And, luckily for us, we don’t have to rely on the Daily Stormer or 8chan to have a pretty good picture of what a site looks like when they’ve gone through something like this. We just need to look at Encyclopedia Dramatica.
Encyclopedia Dramatica was essentially the blueprint for Kiwi Farms. The now-defunct wiki was created by 4chan users to archive internet drama. Because it was created by and for 4chan users, every line on the site was full of hate speech and absolutely inscrutable in-jokes. It also tried to be both an attempt at genuinely explaining early social media controversies and also a meta-parody of Wikipedia itself, which made it borderline unnavigable. It existed for about seven years before it finally went down. Users passed around its archives via torrents for a while after that, but as of 2020, when I last checked on it, the admin who kept mirrors of the site up was in jail and no one else who had archives of it wanted to deal with the stress of trying to run it.
It’s pretty much what always happens to websites like this. And, in many ways, it’s exactly what should happen. Express yourself freely online, but also know that if enough people hate your website and deem it too racist, too violent, or too dangerous to exist, it will eventually become your burden to keep it online. And it turns out that burden is pretty heavy.
There’s A Weekend Edition Of Garbage Day Now
It’s a nice bonus issue of Garbage Day once a week for paying subscribers. If that sounds like something you’re interested in, hit the green button below. You’ll also get Discord access and a bunch of other fun bonus stuff.
Some Good Boston Content
God, the Boston accent is just beautiful, isn’t it? Trying to decide which part I like the best, “swindluhs,” “griftahs,” or the part at the end where Dropkick Murphys lead singer Ken Casey does this very specific Massachusetts thing where he kind of like muffles a laugh into his cheeks sorta. It’s a very unique kind of Boston asshole thing to do after realizing you just got way too serious there for a second. If you know, you know.
Zuck Wants Cool Tough Alpha Chads To Wrestle In His Metaverse
I didn’t watch Mark Zuckerberg’s recent Joe Rogan appearance because I only have one life to live on this cursed planet. But per Bloomberg, Zuckerberg spent a chunk of the interview trying to rebrand himself as a Sports Guy. “He referred to himself as a three-sport varsity athlete, proclaimed a newfound love of jujitsu (he’d hired one of Rogan’s friends as his coach), and, sounding not at all like a billionaire who’d adopted a hobby to impress a talk radio host, said that among his favorite pastimes was ‘just wrestling around with friends. It’s awesome,’” the Bloomberg piece reads.
Then, he released this absolutely ridiculous video of himself training with MMA fighter Khai "The Shadow" Wu. The only physical activity I do is typing posts, so I’m in no position to say whether Zuckerberg is any good at fighting, but I’ve seen a lot of Twitter accounts with names like “@ClvssicTrvdWvrrior,” or whatever, say that Zuckerberg’s footwork is trash.
The timing of Zuckerberg trying to pitch the metaverse as a place for big beefy dudes to bulk up is interesting. Vivek Sharma, the vice president for Horizon, Meta’s main VR platform, is leaving the company, which some took as a bad sign for the company’s metaversal aspirations. Oh, also, the company’s virtual reality division lost $2.8 billion as of last quarter. So that’s probably also bad.
But I think Zuckerberg’s a bit late for courting the Joe Rogan contingent. Most of those guys got super into crypto and now see his platform as the enemy. I, personally, think Zuckerberg should give up on that demo entirely and instead rebrand as a vaguely-goth dark academia Tumblr sexyman and try and see if he can get Horizon to catch on with that crowd instead.
Misinformation Should Not Be The Users’ Responsibility
The tweet above is from Twitter’s Safety team and it’s part of series of infographics the account shared recently. This one is titled, “Accidentally shared misinfo,” and then it lists a bunch of things you can do to fix your mistake. I wrote about this on Friday, but the entire idea of policing yourself from sharing misinformation is ridiculous.
First, I just do no think the average user cares enough about this stuff to do this. Second, I don’t think they should have to! That’s not their job. And, third, imagine literally any other company in another other industry doing this. Can you imagine a company crazy enough to publish something that reads, “Accidentally bought a Big Mac for your children that made them sick? Here’s what you should do next: 1) make your kid throw up 2) buy a new Big Mac that looks safer to eat 3) feel empowered to keep shopping at McDonald’s.”
Sorry, I don’t want to be a boring scold about this, but c’mon.
A Truly Maddening Internet Mystery
I hate that I sort of recognize this. And I absolutely definitely hate that I can’t remember what it’s from. As far as details we have to go on, the photo was taken in 1992 in Southwest Ontario. I spent a lot of time going through the thread over the weekend and, so far, it doesn’t seem like anyone’s figured it out.
Hilariously, someone thought it may have been from a commercial, so they built a neural network that could scan 50 hours of old 90s commercials and couldn’t find anything that matched.
The A.I. Prompts Will Be Replaced With A.I. Next
Benoit Martinez, a director at Ubisoft Paris, has been sharing some experiments he’s done randomizing different variables that can be loaded into A.I. art tools and the results are pretty cool.
If you’ve never used one of these applications before, you basically type a command or prompt into a bar and the A.I. tries to generate what you asked. Users have quickly realized, no pun intended (sorta), that there’s actually a real art to getting the prompt right. This has even led to a marketplace where people can buy and sell good prompts that are guaranteed to get the A.I. to spit out the right thing. There’s also a lot of conversation right now about whether or not a prompt can be copyrighted. It’s an interesting subject all around, but I think most of the handwringing about it won’t really matter in the long run.
In a recent podcast episode I did on A.I. content, I compared A.I. prompt writing to search engine optimization, or SEO. There was a five year period across the late 2000s and early 2010s where a bunch of men (always men) got paid a lot of money because they claimed they knew all the tricks to getting a site to the top of major search engines. They knew how to speak to the machine better than anyone else, they said. These thinkfluencers eventually fell by the wayside and now most media organizations (as well as just most major companies) have a team of people whose job it is to stay current with how to best get their website on various search engines. I sort of assume that the current A.I. prompt cottage industry will end up being a compressed version of that and in like nine months we’ll find out some marketing firm has an A.I. optimization team that’s figured out the right syntactical formula to get DALL-E or Midjourney to iterate a zillion corporate logos per second or something.
In fact, that may be getting closer than we think. Some very funny screenshots from what looks to be an A.I. prompt writer Facebook group were dropped in the Garbage Day Discord by JRo. Seems like users in there really did not like the idea that they might be automated, as well. And, also, there’s literally already a tool called Phraser that can do exactly that. Womp womp.
A Very Good TikTok Account
Enable 3rd party cookies or use another browser
I came across this thanks to the always-excellent Tauriq Moosa on Twitter. A British band called Deco has been doing a great series on TikTok where they’re performing slightly more modern songs in the style of older 80s artists. The one embedded above is “Teenage Dirtbag” performed by Rick Astley. Definitely recommend clicking through and checking them all out. They absolutely rip.
Some Stray Links
Why are straight British men so revved up about London’s new train?
P.S. here’s an incredible Rube Goldberg machine.
***Any typos in this email are on purpose actually***
man, i was so psyched to read kiwi farms was finally kaput...but that statement, haha, wow. thank you for taking the ridiculous phone comparison to task - 2-3 people talking in a private conversation about terrible things is nothing like massive platforms, big yikes. and also, it's weird they spent so much time talking about ~fanning the flames~ when if they really gave a shit and/or were concerned they wouldn't post a statement that is in itself inflammatory, like ??
meanwhile, i'm not sure the big mac analogy landed as strong...actually, i don't really understand it at all. analogies aside, what's wrong with correcting yourself? obviously if you're only tweeting to like, ten friends then who cares i guess but, when you have a larger platform, corrections are important. so much of the internet now is just droves of people champing at the bit to have the next ~hot take~ or funniest dig and then everyone in the comments repeating it (but saying it like "i don't know why but i feel like [repeat of the post/comment they're replying to]") and anyone who corrects them just gets trolled and laughed at. it's so weird. i digress but, to get back on topic - i don't expect anyone to do thesis level research before posting something online. but if you find out you were wrong, why not say so? i feel like i must be missing something.
-the rube goldberg machine was amazing - but so was the camera work to capture all that!
-enjoyable tiktok. i saw wheatus at the height of "teenage dirtbag" and the crowd hated all their other songs so much they played teenage dirtbag twice. or three times maybe.
-dropick murphys forever and ever amen. i'm a rhode islander who has been in boston for over 20 years, and you can hear in this clip something people who try to do our accent never get right. which is that you don't just drop every R ever. he very clearly says both "swindluh" and "swindleR." our accent gets more intense if we are hyped up about something (good or bad) and/or yelling. and yes, we will get all whizzed up and then be like, lol anyways have a good one tell your mom i said hello
You may not remember this from however many years ago, but the language Matthew Prince used in the post directly mirrors the FCC definition of a "Common Carrier", who are among other things held to what is usually referred to "net neutrality", and this is richly ironic considering internet service providers have been resisting this classification for decades.
There is probably an argument to be made that, yes, Cloudflare should be considered a "Common Carrier" in some sense or another, but Matthew Prince clearly wants to have his cake and eat it, too, in a way substantially analogous to Spotify pretending that their business deal with Joe Rogan was somehow covered by Section 230.
More broadly, though, the problem with "free speech" on the internet is that it's a form of misdirection that companies deliberately use to distract people from the largely unaccountable monopolistic power they have amassed. Or, again, the primary function of Section 230 is to help private tech companies establish and maintain monopolistic cartels.