What should Elon do with Twitter?

I was having fun making snarky comments about Elon Musk buying Twitter, but when a friend asked me what I thought Elon should do with it, I had to admit I had no idea. Happily, deciding Twitter’s fate is above my paygrade, but here’s the most plausible strategy I could come up with:

Twitter should become the iconic, permanent, and bulletproof way for people to have public conversations. This would be massively valuable on a human-civilization scale and comes down to three core parts, configured in a self-reinforcing loop and improved over decades.

1. Human-scale the lies
2. Make it indestructible
3. Free the curation

Here’s a sketch of how this could work.

Human-scale the lies

It’s not possible (and probably undesirable) for Twitter to eliminate individual lying, but lying at scale is a different problem. Make it almost impossible to do anything on Twitter without specific human action. No automated posting, retweeting, liking, account creation, etc. Verify that every account is controlled by a human and make it hard for one person to create more than a couple of accounts. Slow all ‘write’ actions down to work at human speed but make automation prohibitively inefficient. When I see something on Twitter, I should be confident that an actual human posted it, and I should be able to see everything else they’ve ever written or shared and haven’t deleted.

Make it indestructible

No person, organization, natural disaster, or rogue government should be able to silence Twitter. Build countermeasures. Tunneling through multiple servers, embedding content in other platforms, transmitting through shortwave radio, and, I dunno, moon lasers. Open-source the receivers so anyone can build a Twitter client out of a soda bottle, two aspirins, and a SIM card. Make it so when the rest of my world goes to shit, I can rely on Twitter to help me get through it. And then to help me rebuild.

Free the curation

The iconic, permanent, and bulletproof way for people to have public conversations will generate a lot of content. Some of it horrible. Most of it dumb. Let people, brands, and governments create curation algorithms that determine what users see. Have the algorithms transparently compete across multiple dimensions: engagement, happiness, revenue, etc. Let users pick which curators they want. Default algorithms can be set in accessible ways, but a different content universe is always just a click away. Other than in exceptional cases, posts and people don’t get blocked by Twitter; they fall out of mainstream curations. Algorithms can be chained, so if I want to add a hate speech filter from one source to a bullshit filter from another, and run the whole thing through my favorite new Netflix show, I’m free to mix and match. Recommendation engines for recommending recommendation engines? Fun!?

These three areas should support each other. Verified users with their entire public history available will make curation easier. Server-side curation and algorithms explicitly geared for emergencies will improve resilience. The flywheel keeps spinning and getting faster.

Doubtlessly many people at Twitter have been working on versions of these three things for years, but it’s time to recommit and eliminate all distractions.

Of course, this punts on the question, “how will Twitter make money?” Nothing outlined above should get in the way of either direct or indirect revenue models, but I think the question is premature. Many of Twitter’s current problems can be traced back to becoming a public company too soon and chasing advertising revenue. That’s too many stakeholders to please before achieving the mission:?

Make Twitter the unfuckable way for people to speak publicly. Advance human civilization. Then figure out the money.

Or Elon can rename it to “The Other Boring Company.” That joke is worth something.?