I've been mulling it over in my head over and over again; what went wrong that could have gone right, and to just what extent the game was rigged from the start.
I have never felt more vindicated in considering animosity, trolling, bad faith, entitlement, and >(ab)users< more generally a threat to Cohost. Obviously Stripe is the bigger factor, but we should still be noting and studying the irony that this place succumbed to the same diseases that took out ye olde snouts.online and entire migrations to fediverse: the losing battle that is trying to maintain a social media website of any size day after day and year after year, when you are surrounded by enemies from within and without for no fucking reason. People who will openly wish harm on you and anticipate the day you fail because you dared to build something good that's not perfect.
Some are more hardened than others. But this is not an experience that the human psyche was built to endure. (Something something panopticon.) It does not surprise me in the least that it took its toll especially rough on people who personally felt the need for this kind of platform, one that aggressively protects its users to the point of restriction (in visibility more than functionality), while pushing back against outside demands for even more restriction, the logical conclusion of which is a platform for nobody.
Consider how things that Jae and Kara, two trans women, have said, especially regarding the immense pressure of being demonized for every decision and every delay, have themselves been picked apart by the mob. Consider the feeling of people you saw as peers, that you are trying to build something for and help and serve and protect, suddenly turn to call you (quoting from an early Cohost opponent) "tankie suicide-baiting pedophiles" because you didn't think about one small part of the content policy hard enough and decided not to ban an edgy slogan against Hitler and his fans- even if you were wrong.
Study the autopsy of a death by toxicity. Because...
It will kill again.
It will kill small FOSS projects. It will kill big FOSS projects. It will kill Fediverse instances. It will kill projects like this one. It will kill organizations and friend groups that have nothing to do with software or hardware or antisoftware.
There's no way to stop it, not really. All we can do is acknowledge its existence and insulate ourselves from its material repercussions and overly optimistic emotional expectations of The Public or our naivete regarding "maybe it'll go away" or how impossible it is to deal with. All we can do is build ourselves stronger and more stubborn, and build what we make more resistant to schisms, attacks, naysayers and other nonsense, and not build things that rely on everyone ever believing in them. If we can build something that somehow never generates any controversy directed towards the creators, all the better, but I don't count on that happening.
The social media Cohost tried to create existed under the main following specifications:
- Design that impedes use that is harmful to its users, and thus technically less powerful, discoverable, and "useful", while increasing meaningful engagement
- Strong moderation and content/etc policies filling in the gaps, keeping the site squeaky-clean
- fun :)
In execution:
- The site mainly appealed to a vocal minority of largely queer, Twitter-burned harassment survivors whose expectations for "a good site" were, in a few cases, maybe a tad unrealistic
- The content policy was "wrong" for a few hours and we've never heard the end of it. People immediately left the site and never came back because the content policy was "wrong" for a few hours, and that meant that the mods loved pedophilia or whatever. We had an entire row over the all-white (oh God, if I'm wrong here) staff being too permissive of racist microaggressions. I could make another entire post out of this bullet point but I'll spare you. You get it.
- broke :(
It's true that the design was a strength and a weakness, whereas the criticism of moderation and policy was pretty much universally along the lines of "not strong enough", but the thing is, it would have been impossible to meet every demand forever. The first few that turned into big conflicts, sure; then people get adventurous. The best example of this is in communities that have congregated around opposition to X type of internet content or whatever, and the growing extent of what they consider to be X and therefore verboten. You can say that the line should be drawn over there instead of over here, but moving the line forever is just not sustainable.
In the grand scheme of things- and people really don't like talking about this one!- Cohost's government was much, MUCH more on the "over there" side of things, in a way that drastically limited its mass appeal. This would have been fine, if Cohost was just a sideshow for Eggbux like the original plan was, as Cohost didn't directly generate revenue from mass appeal and gravity via ads. If you try to build Cohost around mass appeal and ads (not necessarily ruling out their inclusion on the side), eventually you have to give up everything that makes Cohost what it is in pursuit of impressions, and you just end up with Tumblr with software that is better in some ways and worse in others. So that's obviously a no-go.
Things we could (not "should!") try the next time we build social media, most of which sacrifice major points of appeal for Cohost:
- Way, way, way, way, way lower expectations from mods. I don't know how to codify this without letting the site turn into a wild west like Bluesky or Tumblr, though I guess Cohost's design is a bit less cloutseek-y/rewarding of controversy than Tumblr's, and maybe a bit more private in a way I haven't thought of. Demographics and inclination of the userbase can't be the whole equation- we saw this when the Tumblr folk blew in and brought some different ideas with them. It has to be design. I think the presence of visible, personable moderators and administration is the problem, because then the userbase can villainize them. So I think this just isn't possible.
- Conversely, replace tactically limited design with more, and more aggressive, moderation. But this doesn't scale the same, which means you need more money despite broader appeal, which kind of self-cancels and leads to...
- Some kind of freemium model. Seen with app.net (which Cohost actually outlasted as an actively developed site and not essentially a zombie- apparently subscription retention was terrible), micro.blog (which was semi-federated at the start, as it used RSS to listen to other blogs, and now it's more of a Fediverse concept), write.as (which has always been Fediverse), and others. This at least addresses the complaint people had that Cohost plus! didn't offer enough?
- Federation. This would make the problem of administration and moderation better in some ways, and worse in others. This was created as an alternative to federation, so seems a bit silly.
- Not advertising as a safe space so as not to become a gathering place for high-maintenance users. The entire design of the site is built around this, so if staff didn't, the community would find it and do it for them. It would have to not be designed as a safe space, which leads to the first two. This somewhat happened to Pillowfort, which never really got off the ground. (Pillowfort is also a good example of "just scale the site down and run it on the side!"- you're describing Pillowfort, which is not what a lot of people came here for.)
- Detached staff that are paid well and don't actually care about the site's userbase. You have to attain gravity before you can have this attitude, though. Not even profitability can save you if you don't have the gravity to keep the users you piss off from leaving. Which is obviously a toxic relationship.
- Do the exact same thing again, but without the investment money. No.
- Do the exact same thing again, but with $100M. Purely hypothetical, of course. If one megarich bastard magically decided they wanted to pour it all into social media and didn't care if it ever made money, I don't see why it wouldn't last a lot longer. But at that point, just make Eggbux, which would make the platform sustainable... and give everyone a pony, while you're at it. The creators of such a site could get flamed to high heaven and it wouldn't matter because they would be making a tidy profit and paying bills. Personally, and this may be a bit of a hot take, I don't think waiting for Bill Gates to personally come save the internet is a good strategy. We're just leftoid Musk stans at that rate... right down to the focus on the payment stuff.
- Do the exact same thing again, but with a viable payment processor. If only, eh?
If you couldn't tell, my least unfavorite of these is the freemium model. Not only does it resolve a lot of the central conflict at the crux of Cohost's concept, but it also has the monetization solution built in, at least in theory.
You also probably need to step into the cheap web hosting space, as the people who have gotten into these freemium services have been disproportionately into the writing and blogging in comparison to regular social media users. That means support for custom domain names and fully customizable pages.
I'm envisioning a service halfway between micro.blog, write.as, WordPress and Tumblr. Free tier users would be able to like, share, and comment on blogs, and make microblogs on a simple profile page, with the specifics of the limitations on a microblog to be determined. Paid tier users would be able to make long blog posts and post larger media types than free tier users, as well as manage a fully customized webpage at a custom domain name. CSS crimes would become the domain of the premium account. No idea how, for example, QRBing something long with something short would factor in.
In practice, the restrictions would be pretty off-putting to a lot of incoming users. Add to that the Cohost philosophy on overexposure/discoverability. This would be a good platform for the "webrings" concept that was discussed and never implemented. Here, this would just make the site feel like Pillowfort, but worse for free users and better for paid users. Presumably better software and "new website energy" is not going to make that big a difference. There's already a lot of overlap between the Cohost demographic and the Pillowfort userbase, even if it would be more if it didn't spiral into a snowball effect with the dwindling userbase on account of the dwindling userbase.
Ultimately though, it feels like I keep making the case for Pillowfort. If you know me at all, you know that I don't know how to feel about this. I've also already halfway made my pivot to the Fediverse and Bluesky. I suppose I could just head over to PF and preemptively block a bunch of people. Are we all headed to Pillowfort? Are we gonna make it not dead? That'd be something... They could use the userbase, and their platform is more similar to ours than the other alternatives most people mention... Who am I gonna follow? Longposters? Anyone?...
I never thought I'd say it, but I'm starting to believe in the Fediverse again. I see a future where the longposters run their own one-person WriteFreely or whatever software stack instances and the rest of us eat that shit up on our own softwares. It seems like this over Pillowfort is the direction a lot of the longposters are drifting in, with others imagining they'll quit social media. (Which is their prerogative, but maybe they haven't considered all the options.) Speaking for myself, I've learned a lot from this website about how not to get tilted at stuff. It feels a lot more livable over there than it used to... it's still not that active, and it's got tons of its own problems, but, well, them's breaks.
In my wildest dreams, this is just a temporary setback; somewhere along the way, an appropriate payment processor will be developed, and Cohost will pick right back up where it left off, complete with Eggbux.