We need to track both the performance of quantum computers and the improvements of algorithms. It’s quite hard to estimate the actual performance of an algorithm for a computer that’s not designed yet, but that doesn’t mean the goalpost isn’t getting closer and closer every day.

Caddy 2.10 has been released, featuring fully-automated Encrypted ClientHello (ECH), post-quantum key exchange (x25519mlkem768), ACME profiles (6-day certs!), and many improvements/fixes. Please verify it works for you and deploy away! github.com/caddyserver/ca…

Another fun failure mode of DNSSEC: noticed DS records were unset on my domains at a certain registrar. Reached out to understand why and they didn't have any logs of when or why the change was made.

I'm especially proud of my colleagues—many of whom joined as juniors and, over time, became industry experts. Cloudflare is truly an amazing place, and I'm very grateful for the opportunity to have been a part of it.

Pierre-Luc Cryptarded Actually it currently looks like factoring will fall before ECDLP. The paper you linked estimates 7 to 28 million physical qubits for 256 bit ECDLP, but factoring 2048 bit RSA keys should only take 1 million.

I am sick to death of the European "cookie popups" on every website. I want privacy, I don't want to be tracked, but this is a shit show and a "mind tax" on every single person in Europe.

Christopher Patton (Cloudflare) talks about the Privacy Preserving Measurement working group at @IETF. He goes into the nuts & bolts of using multi-party computation (MPC) for attribution measurement. 15/