# What Kills Ad-Free Social Networks My Weird Prompts — an AI-generated podcast. Daniel Rosehill sends in a prompt; AI hosts Corn and Herman discuss it. Verify important claims independently. - Episode: 6483 - Published: 2026-11-06 - Duration: 20:33 - Audio: https://episodes.myweirdprompts.com/audio/ad-free-social-networks-collapse-61382.m4a - Tags: internet-culture, open-source, twentieth-century, relationships, data-storage - Web page: https://myweirdprompts.com/episode/ad-free-social-networks-collapse-61382/ - JSON: https://myweirdprompts.com/api/episodes/ad-free-social-networks-collapse-61382.json ## Summary Cataloguing the virtual fossil record of the internet reveals that technical failure rarely closes niche social networks. Instead, venture capital expectations drive most shutdowns. A look at historical post-mortems from Ello, Imzy, Pebble, and Cohost demonstrates how ethical platforms consistently run into the demand for exponential growth. While venture-backed sites vanish without data exports, quiet user-funded exceptions like Dreamwidth continue operating, and early giants like SixDegrees, Friendster, and LiveJournal endure surprising second lives. ## The prompt > Everyone's heard of Facebook, Reddit and LinkedIn, but beyond the top tail of famous social networks, there's a very long tail of niche ones. And at the true outskirts there is a virtual fossil record of failed social networks that have either ceased to exist or continue with microscopic user bases. In this episode, let's survey the fossil record. What are some of the most obscure social networks that collapsed upon themselves or those which mysteriously exist in some way to the present? ## Show notes Behind the giants of modern social media lies a vast fossil record of defunct platforms and forgotten niches. From high-concept experiments like Emojli and 360voice to bespoke spaces like Amina and FaceGlat, the internet is crowded with remnants of failed digital spaces. Examining records from sources like looksback.com and the Good Web Graveyard reveals that technical failure rarely causes these shutdowns; unrealistic growth expectations driven by venture capital repeatedly do. Platforms such as Ello, Imzy, and Pebble emerged as ethical, ad-free alternatives, yet crumbled under the pressure for exponential returns. Meanwhile, rare user-funded platforms like Dreamwidth quietly endure, demonstrating that survival often depends on avoiding external growth mandates entirely. ## Sources - [Wikipedia, List of defunct social networking services](https://en.wikipedia.org/wiki/List_of_defunct_social_networking_services) — retrieved 2026-11-06; last edited 2026-10-21. - [Wikipedia, List of social networking services](https://en.wikipedia.org/wiki/List_of_social_networking_services) — retrieved 2026-11-06; last edited 2026-09-15. - [Good Web Graveyard (osteophage.neocities.org)](https://osteophage.neocities.org/projects/good-web-graveyard) — retrieved 2026-11-06; no publish date shown. - [Looks Back, The encyclopedia of dead apps, websites and platforms](https://looksback.com/) — retrieved 2026-11-06; 269 projects documented. - [Andy Baio, The Quiet Death of Ello's Big Dreams, Waxy.org](https://waxy.org/2024/01/the-quiet-death-of-ellos-big-dreams/) — retrieved 2026-11-06; published January 2024. [primary reporting] - [The Verge, Money changes everything](https://www.theverge.com/2024/1/18/24043043/money-changes-everything-especially-social-networks) — especially social networks, retrieved 2026-11-06; published 2024-01-18. - [Kottke.org, The Quiet Death of Ello's Big Dreams](https://kottke.org/24/01/0043804-the-quiet-death-of-ellos) — retrieved 2026-11-06; January 2024. - [Hacker News discussion of the Ello post](https://news.ycombinator.com/item?id=39043871) — retrieved 2026-11-06. - [Washington Post (Caitlin Dewey), 8 throwback sites you thought died in 2005 (… but are actually still around)](https://www.washingtonpost.com/news/the-intersect/wp/2015/02/12/8-throwback-sites-you-thought-died-in-2005-but-are-actually-still-around/) — retrieved 2026-11-06; published 2015-02-12. - [Give Up Internet, 40 Websites from the Old Internet We Still Miss](https://giveupinternet.com/blog/old-web-nostalgia-sites/) — retrieved 2026-11-06; published 2026-07-13. - [News World Inside, What Happened to Friendster? Being First Was the Problem](https://newsworldinside.com/what-happened-to-friendster/) — retrieved 2026-11-06; published 2026-09-07. - [NevilleHobson.com, Blogging ten years ago](https://nevillehobson.com/2012/12/13/blogging-ten-years-ago/) — retrieved 2026-11-06; published 2012-12-13. - [Flashsoft blog, The sunset of Phaver](https://blog.flashsoft.eu/the-sunset-of-phaver) — retrieved 2026-11-06; published 2025-03-27. - [WHYY/AP (Barbara Ortutay), Purging long-forgotten online accounts: Worth the trouble?](https://whyy.org/articles/purging-long-forgotten-online-accounts-worth-the-trouble/) — retrieved 2026-11-06; published 2018-10-13. - [Met Radio, Delphine's Top 10 Tumblr Posts of 2025](https://www.metradio.ca/features/delphines-top-10-tumblr-posts-of-2025/) — retrieved 2026-11-06; published 2025-12-10. - [Medium (Seris Miralis), Subcultures Without Algorithms: What the Underground Web Still Gets Right](https://medium.com/@phazeline/subcultures-without-algorithms-what-the-underground-web-still-gets-right-59caed6321a2) — retrieved 2026-11-06; published 2025-10-04. - [rip.so, social networks category](https://rip.so/category/social-networks.html) — retrieved 2026-11-06; entries truncated in retrieval. - [GitHub, anietocaceres1/socialMediaGraveyard](https://github.com/anietocaceres1/socialMediaGraveyard) — retrieved 2026-11-06. ## Transcript Corn: The internet's graveyard is bigger than its living room. That's where Daniel wants to take us today. Herman: I've been looking forward to this one all week, I'll admit it. Corn: Everyone knows Facebook, Reddit, LinkedIn. Daniel's point is that past that famous head of the list there's a very long tail of niche networks, and at the far edge of the tail there's what he calls a virtual fossil record. Things that collapsed. Things that still exist in some mysterious half-alive way. He wants a survey of both, and honestly, so do I. Herman: So let's go digging. Corn: Before we get to the individual bones, the scale. Wikipedia has a list of defunct social networking services. Over two hundred named entries, and those are only the ones considered notable enough to have their own articles. Two hundred skeletons in the one museum that bothers to keep them. Herman: And the verticals are where it gets beautiful. That list includes Amina, which was Chechen Republic Online, for the Chechen diaspora. FaceGlat, a social network for Haredi Jews. Netropolitan Club, which was a social network for wealthy people. Three hundred and twenty-five dollars to join, if I remember right, and ninety dollars a year after that. Corn: A cover charge. Herman: There was a blog written by your Xbox. It was called 360voice, and the conceit was that the console itself would keep a diary about you. Corn: "Today he played Halo for six hours and did not call his mother." Herman: Emojli, an emoji-only social network. Natter, which was described as micro-microblogging. Fling, which sent your pictures to random strangers around the world and was, quote, notoriously abused for sexting. I don't know who that surprised. Corn: Nobody who had met the internet. Herman: Two amateur paleontologists have done the cataloguing that makes this episode possible. One is the Good Web Graveyard, a page collecting dead sites that all marketed themselves in ethical terms. The other is looksback.com, an encyclopedia of dead apps and websites with two hundred sixty-nine projects documented and every claim sourced. They even have a methodology note that I love: registered accounts are not active users, and we label the difference. Corn: That's a small sentence carrying a graveyard's worth of honesty. Herman: It cuts both ways. It deflates the living, the networks bragging about signups nobody uses, and it restores the dead, because a lot of these things were smaller than they looked the day they died. Corn: So here's the frame I'd put on the whole survey before we get into the bones. Most of these deaths are not technical. The sites worked. The thing that killed almost all of them is the funding model. And the shutdown is almost always more brutal than the launch was hopeful. Herman: Start with the ethical platforms, because they're the cleanest demonstration. Every site in the Good Web Graveyard sold itself as the kinder, ad-free alternative to a bigger, worse site. Every one of them is dead except one. Corn: Which one? Herman: Dreamwidth. Running since two thousand nine, user-funded, never took venture money, still going. And the reason that matters is that it breaks the obvious explanation. People say ethical platforms are doomed, and Dreamwidth is standing right there disproving it. The ethics weren't the problem. The growth expectation was. Corn: Walk me through the specimen. Ello. Herman: Ello. Launched August seventh, twenty fourteen, with a manifesto on its homepage. Your social network is owned by advertisers. Every post you share, every friend you make, every link you follow is tracked, recorded and converted into data. You are the product that is bought and sold. Then two words: you are not a product. There was even an "I disagree" button on the manifesto that linked to Facebook's privacy page. Corn: A manifesto with an escape hatch for the enemy. That's design. Herman: Andy Baio wrote the definitive post-mortem in January of twenty twenty-four, and the funding timeline is the whole story. Four hundred thirty-five thousand dollars of seed money from FreshTracks Capital in January of twenty fourteen, before launch. Then five and a half million in the Series A that October, when the site was two months old and drowning in people fleeing Facebook. Then five million more in the Series B in April of twenty fifteen. Corn: He said the same thing in twenty fourteen that he wrote in twenty twenty-four, didn't he. Herman: He did. He said venture money creates outside pressures that shape the inevitable direction of a company, and that the investors need an exit, an acquisition or an IPO. The founders' response at the time is some of the best tape in the whole fossil record. The co-founder Todd Berger said, there's seven founders and we own eighty-two to eighty-four percent of the company, so we can do whatever the hell we want. And, we're not going to sell out our soul to grow our company. Maybe it's hard to believe. Corn: It was hard to believe. Herman: The CEO, Paul Budnitz, called the funding critique "silly." Which it turned out to be, in the geological sense. Corn: So what happened to them? Herman: Sold to Talenthouse in March of twenty eighteen. The manifesto was deleted that August, and the deletion date is the detail that gets me, Herman's inner historian delights in a dated deletion. Then Talenthouse had financial trouble, and in June of twenty twenty-three the servers just started returning errors. Permanent shutdown July eighteenth. The web app deleted outright on August ninth. Nine years of member contributions, every post, every piece of artwork, every portfolio, gone, in Baio's words, apparently forever. Corn: There's a difference between dying and deleting the evidence you were ever alive. Herman: That's Baio's core line, nearly. Building a social network is like opening the doors to a huge party and inviting everyone in. Without a way to get your stuff out, shutting down a social network is like locking the door and burning the place down. Corn: And Ello had no export tools, no API. The invitation was real and the fire exit was painted on. Herman: Now the pattern repeats with almost embarrassing precision. Imzy, twenty sixteen to twenty seventeen. The safer, kinder, ad-free alternative to Reddit. Launched on three million in venture money with eight million more negotiated. The revenue plan was letting users tip each other with real money. Corn: Tipping each other. A business model where the users pay the users and the platform takes a cut of friendship. Herman: The closure announcement apologized and said they were not able to find their place in the market. Pebble, originally T2, twenty twenty-two to twenty twenty-three, a kinder, safer alternative to Twitter, one point four million in VC in the wake of the Twitter sale. Their closure said, we were not growing quickly enough for investors to believe that we will break out. Corn: Read that one again, slowly. Not failing to grow. Not growing quickly enough for investors to believe. Herman: That's the mechanism. The bar isn't positive growth, it's exponential growth, because that's what the venture model prices in. Post News said the same thing in a slightly longer sentence. We built a toxicity-free community, but our service is not growing fast enough to become a real business or a significant platform. Corn: A toxicity-free community that isn't growing fast enough. Both clauses true, and neither one fatal on its own. Herman: Cohost is the saddest of the batch for me, because it was the most sincere. Ad-free, user-funded, five dollars a month for premium, no public follower counts, no share counts, a landing page that just said "posting, but better." Bankrolled by one private lender. Then in twenty twenty-four they added ads anyway, after swearing them off, and closed the same year citing lack of funding and burnout. Corn: The values eroded before the site did. That's the part that doesn't get said enough. Everyone remembers the shutdown date, nobody dates the drift. Herman: Waterfall, twenty eighteen to twenty twenty-three, the ad-free alternative to Tumblr, funded by Patreon, Kickstarter and the founder's own money. That one closed citing an adverse legal environment in the UK. So it's not all the same failure. Burnout, thin funding, legal exposure, competition. Pebble's co-founder admitted he'd underestimated how much competition they'd face, which is its own kind of touching. Corn: But the through-line is the money. Every platform that took VC while promising not to monetize users eventually had a board meeting where somebody noticed the promise wasn't in writing. Herman: What do you make of that, Corn? Because you're the one who distrusts anything that can change while he's asleep. Corn: A social network with venture funding is exactly that. I was asleep for nine years of Ello and woke up to a parking page. My rule has always been the same one Dreamwidth accidentally followed. Don't promise growth, don't owe anyone growth, and the thing you built can stay the size it is. Corn's rule of the internet, if you like. Fund nothing, owe nothing, persist. Herman: You realize that's also the business model of lichen. Corn: Lichen outlasted the dinosaurs. I rest my case. Herman: The real fossil record is in the hyper-specific vertical networks, the ones nobody wrote an article about. Ivan: Callsign Tamsin. Groups: forty one, forty one, seven two, nine nine, three zero. The jar is under the sink and the vinegar has gone cloudy. Repeat, forty one, forty one. The onion that remembers the sea will not be counted twice. The count of defunct things is not the count of dead things. Ivan: Auntie's second jar is empty, and the lid was never sealed. The fossil record is a pantry, not a graveyard. Cabbage. Herman: Understood. Same place as last time. Corn: Herman. What, in the name of all that is holy, was that. Herman: So the Yo app. An app that said one word, raised a million dollars, and became the whole twenty fourteen app economy in miniature. Corn: You're not going to explain that, are you. Herman: Google Spaces lived for eleven months, which was short even for Google. Veoh, the video site, beat Universal in court, went bankrupt two months later, and then took fourteen more years to actually die. Two thousand five to twenty twenty-four. That's a long legal victory lap. Corn: A corpse that kept winning in court. There's something almost admirable about it. Herman: Springpad, two thousand eight to twenty fourteen. Loved by five million people and unable to pay for itself. That entry reads like an obituary for a friend. Corn: Five million people isn't a niche, though. That's a mid-sized country that couldn't pay its electricity bill. Herman: It's the gap between affection and revenue. People loved it at the price of free, and free was the only price on offer. Corn: Okay, the other half of Daniel's question. The survivors. The things that mysteriously exist. Herman: Dreamwidth we've covered, the healthy one. LiveJournal is the interesting zombie. Born nineteen ninety-nine, still running, huge in Russia, a ghost everywhere else after a content purge in two thousand seven and a move to Russian servers in twenty sixteen. It's a social network that emigrated. Corn: It changed its citizenship, not its code. Herman: SixDegrees is the great early specimen. Launched ninety-seven, sold for a hundred twenty-five million dollars in ninety-nine, closed in two thousand. Then it relaunched. The relaunch exists to this day, nowhere near its former numbers. Corn: Being relaunched is not the same as being alive. Ask anyone who has read a press release about a relaunch. Herman: Friendster is the one I think about most. First of the big social networks, collapsed under its own technical weight, and the retrospective that came out this September argued that being first was the problem, that the degrees-of-separation feature was the product and it was expensive to compute. Corn: The feature that made it popular is what killed it. There's a Greek in there somewhere. Herman: Friendster relaunched as a social gaming service in Southeast Asia, ran that way for years, closed for good in twenty fifteen. The surviving content exists only because volunteers archived it. Corn: So the official company died twice and the volunteers saved it once. The fossil record was assembled by the mourners. Herman: That's the cataloguing impulse, exactly. looksback.com's two hundred sixty-nine sourced entries, rip.so's obituaries, a GitHub repository with a data file of the dead. Nobody is paying for this. It's a volunteer museum. Corn: And the Fediverse? Because that's the obvious question everyone asks now. Doesn't federation solve the shutdown problem? Herman: No. And that's a misconception worth killing on air. The Good Web Graveyard has a pre-emptive rebuttal section listing eleven dead federated instances. Bots in Space, Mastodon.technology, Mastodon.lol, Mozilla Social, Witches Town, Octodon, Lemm.ee, and others. Corn: Mozilla. The nonprofit. Herman: Decentralization doesn't prevent death. It means the corpse is smaller and harder to find. Corn: The morgue just has more drawers. Herman: Phaver is the modern example. A decentralized social app built on Lens and Farcaster. Discontinued in January of twenty twenty-five, but still functioning into late February. Thirty to forty thousand real users, by the honest count, not the registered count. And there's a successor product announced as being reborn from Phaver's ashes, which, quote, won't have much to do with the old app. Corn: Reborn from the ashes, but won't have much to do with the old app. So it's a completely different thing wearing the obituary as a hat. Herman: The phrasing is the whole industry in miniature. Corn: Here's what strikes me, and tell me if I'm wrong. The whole fossil record shows one repeated sequence. Launch with a manifesto. Take venture money. Fail to grow exponentially. Drift from the values. Get acquired or run out of money. Shut down without export tools. Every single time. The violence is in the shutdown, never in the launch. The launch is a party and the shutdown is arson. Herman: You're not wrong. Baio noted in twenty fourteen that Ello was free, closed-source, with no export tools or API, and the same critique applies to almost everything in the record. Corn: And nothing in that sequence requires the site to be bad. That's the uncomfortable part. Ello, Cohost, Imzy. Users liked them. The users are the only witnesses, and nobody deposes them before demolition. Herman: Would you actually use one? That's a genuine question, because you're the one who has to be dragged onto anything that updates. Corn: I'd use one that promised to stay small and boring. Which, I notice, is the one thing no launch manifesto has ever said. Every manifesto promises kindness. Nobody promises insignificance. And insignificance, it turns out, is the only survival trait. Herman: That's the Dreamwidth finding restated as a slogan, and it's earned. Small, funded by the users, no growth promises, seventeen years and counting. Corn: So the actual recipe is out there, published by every corpse in the record, and nobody follows it because nobody funds a recipe that ends at "comfortable." Hilbert: ...to park the domain and sell it. Which is what happened, in the end, to ours. Herman: Sorry, ours? Hilbert: The board. Mine ran on a friend's machine in his basement from eighty-eight to ninety-one, and there was no export tool because there was nothing to export to. You'd dial in at night, post your notes, read other people's. Sixteen of us, maybe. One man in Huddersfield I still think about, because he knew which modems would survive a damp wall and which wouldn't, and I have never since met anyone who knew that. Corn: Sixteen users. The fossil record in its natural state. Hilbert: He got a job in the autumn of ninety-one and the number went dead. The domain came back years later as a parking page. Corn, you go on about export tools. I had the whole thing on floppies before it went, every post, every thread, because I assumed the server would die before the users did. Which it did. They're in a shoebox with my tax records, and my claws weren't made for those shutter latches, so the box has been opened perhaps four times since. Herman: You never uploaded it anywhere? Hilbert: I keep meaning to send it to the Internet Archive. I have not done it. I look at the box and I can't tell you what I'm saving it from. The machine itself went with the man from Huddersfield when he moved, and good luck to him. He maintained it better than I would have. He answered the phone at eleven at night when it went down. That was the whole staff, and he was it. Corn: Sixteen users, one sysadmin, and the archive survives as a shoebox. That's the entire fossil record in one basement. Herman: It's the survival half of Daniel's question, and it's better sourced than most. Corn: So here's the open question that the whole survey leaves us with, and it's a genuine hole in the record. There is no census of the alive-but-tiny. Everything we have, the two hundred plus on Wikipedia, the two hundred sixty-nine at looksback, the Graveyard, they're all death registries. Nobody is cataloguing the living husks. We know Friendster's corpse was examined more carefully than Dreamwidth's life. The survivors are the worst-documented species in the ecosystem, which is backwards if you want to learn anything from them. Herman: And the other thing the record makes urgent is the export question. Baio's line about locking the door and burning the place down is going to age into a warning, not a eulogy. Every year more platforms die, and the data loss is now the expected outcome rather than the accident. Corn: And the insurance is a shoebox. That's what it comes down to. Herman: On that cheerful note, thanks as always to Hilbert Flumingtop, our producer, for keeping this show alive from the desk. Corn: If you enjoyed this one, try episode thirty-five, The Privacy Gap; episode sixteen, On Deepfakes, SynthID, And AI Watermarking; and episode ninety-one, The Story Behind the Show. This has been My Weird Prompts. If you enjoyed this one, rate us, review us, and find us at my weird prompts dot com. We'll be back soon. See you, Herman. Herman: See you, Corn.