Every social network you've loved got worse the same way, on the same schedule. It's not the users. It's what the door costs to walk through.
Every social network you've loved has gotten worse. Not all at once. It happens the way a neighborhood declines: the feed fills with noise, the people who made it good post less, then leave quietly, and their absence makes the next departure easier. You didn't imagine it. The decline is real, it's predictable, and it has almost nothing to do with the users everyone blames.
This pattern is old enough that veterans of the earliest internet communities had a name for the month it destroyed theirs, back when a yearly flood of newcomers arrived faster than the culture could absorb them. The flood eventually became permanent, and the culture never recovered. Researchers who study communities describe the same spiral: as average quality drops, the most valuable contributors are the first to leave, which drops the average further, which drives out the next tier. The network cools from the top.
The comfortable diagnosis, and why it's wrong
The standard diagnosis is that bad users arrived. That's the comfortable story, because it means the design was fine and the people were the problem. I think the diagnosis is wrong, and the proof is that the same people behave completely differently in different rooms. The colleague who is thoughtful in a meeting is feral in a quote-tweet. It's not that networks attract worse humans over time. It's that networks price entry at zero, and zero-cost entry makes everyone's laziest, loudest, most extractive behavior free.
What free entry actually buys
Look at what free entry actually buys a platform: growth at all costs. Every barrier removed adds signups, so barriers keep falling until an account costs nothing to create, nothing to abandon, and nothing to replace after a ban. That's rational if your business model needs bodies. But it wires a specific fate into the product, because the value of exploiting a free account rises as the network grows, while the cost of exploiting it stays at zero. Spam, engagement farming, outrage bait, drive-by cruelty: all of it gets more profitable with scale and never gets more expensive. Open networks don't decay despite their success. They decay because of it.
Open networks don't decay despite their success. They decay because of it.
Why gates mostly fail
The obvious response is a gate, and gates have been tried. Invite-only launches, application queues, waitlists. Mostly they fail, and it's worth being honest about why: an invite is usually a ticket. It gets checked at the door and forgotten. The moment you're inside, the person who invited you bears nothing for what you do next, so the gate filters exactly once and then the network is open in every way that matters. A ticket changes who arrives first. It doesn't change how anyone behaves.
## A stake, not a ticket
What changes behavior is a stake. On Chirpper, every member is vouched in by an existing member, and the relationship doesn't dissolve at the door. Reputation flows both ways through that chain for as long as the account exists. Vouch someone in and part of your standing rides on them, permanently. That's the small price paid upfront: not money, but a durable claim on the judgment of someone who already belongs. It costs the newcomer some care in how they act and costs the inviter some care in whom they choose. Neither cost is heavy. Both are constant, and constant is what free entry can never be.
Not money, but a durable claim on the judgment of someone who already belongs.
The spiral runs in reverse
Now run the growth math again with stakes instead of tickets. As the network becomes more valuable, a free account becomes more worth abusing, but a vouch becomes more worth protecting. The same force that degrades an open network tightens a staked one. People spend their vouches more carefully as membership means more, so growth raises the average instead of diluting it. The spiral runs in reverse: quality attracts careful vouching, careful vouching preserves quality.
## The cost, plainly stated
I'll concede the cost plainly, because it's real: this caps how fast a network can grow. Chirpper cannot add a million users in a weekend, and no version of this design ever could. Slower growth isn't a bug I'm reframing. It's the price of the mechanism, and I'd argue the last two decades are one long demonstration of what the alternative price looks like. Every platform that chose speed got scale first and decay second, on schedule.
Where this leaves the industry
Networks don't get worse because people are bad. They get worse because the door is free and nobody inside is responsible for it. Charge a small, human price at the door, keep everyone connected to what they let in, and time starts working for the network instead of against it.