I have tremendous respect for the IETF. They built the protocols that made the internet possible - not as a walled garden owned by one company, but as open infrastructure anyone could use. That achievement is real and shouldn't be diminished.

But I've been thinking about what happens to revolutionary ideas when they enter rooms full of reasonable people trying to reach consensus.

Here's how it usually goes:

Someone arrives with a radical protocol. It does something genuinely new - breaks with how things have always been done. The room full of engineers looks at it carefully. They're smart, well-intentioned people. They start asking reasonable questions.

"What about backwards compatibility?"

"How does this scale?"

"What's the migration path?"

"Can we make this work with existing infrastructure?"

All reasonable questions. All technically sound. And one by one, the sharp edges get sanded down. The revolutionary feature becomes an optional extension. The thing that made it dangerous becomes a "future consideration". By the time consensus is reached, you have something that works beautifully within the existing system - which was exactly what you were trying to escape.

The engineers aren't villains. The process isn't malicious. But the room itself has gravity. And that gravity pulls toward the center, toward compatibility, toward what the largest players can agree to implement.

The companies that can afford to send engineers to every meeting, for years, effectively set the boundaries of what's "reasonable". Not through conspiracy, but through simple presence. They're in the room when the early discussions happen. They're there for the working group formation. They shape the problem statement before most people even know there's a working group.

We've watched this happen. A protocol that could have changed power dynamics gets refined into something that serves the powerful more efficiently.

The dangerous thing isn't that IETF is broken. It's that the process works exactly as designed - building consensus among people who show up. When the people who can afford to show up work for trillion-dollar companies, consensus naturally serves trillion-dollar interests.

This doesn't mean standardization is bad. Interoperability matters. Shared protocols create network effects that benefit everyone. Sometimes you need to be in the room.

But keep in mind - the choice to NOT standardize can be just as strategic as the choice to standardize.

Some tools need to exist outside consensus. Some ideas are only powerful because they refuse to be reasonable. Some protocols win precisely because they don't wait for permission from a room full of experts.

The cypherpunks understood this. They didn't ask IETF to bless PGP. They just shipped it. When governments tried to stop them, they printed the source code in books and mailed it across borders. They treated code as speech and distribution as civil disobedience.

There's a lesson there about the difference between building with institutions and building outside them.

So if you're going into a standards body - go with clear eyes. Know what you're willing to compromise and what you're not. Understand that "open process" doesn't mean "neutral process". Watch who shows up, who stays, and whose "technical concerns" align suspiciously well with their business model.

And remember: some of the most important technologies in history were never standardized at all. They just worked, and people used them, and that was enough.

The internet we have came from IETF standards. But the internet we might still build - the one that resists capture, that empowers small players, that genuinely decentralizes power - that one might need to be born somewhere else.

Not in opposition to standards bodies. Not in ignorance of them. But with a clear understanding that legitimacy and liberation are not the same thing, and sometimes you have to choose.