This is the single smartest thing anybody who's into crypto has ever said to me about it.

It got me thinking, and in fact it got me thinking so hard I wrote 3,500 words about crypto: seldo.com/posts/crypto-the-g… TLDR: I think crypto *will* change the world, for the better. But not in the way its proponents believe.

Thought on the "Transaction costs and abuse at scale" section: something that worries me is that with these ambitious proposed crypto social software systems you need to get the design exactly right up front, because you can't iterate on the smart contracts later!

Can't iterate? Hm. Counter-point: uniswap v1, v2, v3. Not to mention the Cambrian explosion of similar defi contracts. And that's just the "let this one die and migrate to the next" approach, not even considering contracts that hot-swap code without dying.

Hot swappable contract approach, to wit: github.com/Agoric/agoric-sdk…

So that's a mechanism whereby users get to vote on upgrades to the code of the system? I don't think that's compatible with iterative development on non-financial software. Imagine if Twitter had to put every little change they wanted to try up for a vote!