techcrunch.com/2025/09/05/x... shouldn't surprise anyone but quotes me so it's obviously good
05.09.2025 16:53 โ ๐ 20 ๐ 4 ๐ฌ 0 ๐ 0@superfell.bsky.social
https://github.com/superfell https://hachyderm.io/@superfell
techcrunch.com/2025/09/05/x... shouldn't surprise anyone but quotes me so it's obviously good
05.09.2025 16:53 โ ๐ 20 ๐ 4 ๐ฌ 0 ๐ 02 images from hackers as characters converse. "I've got a record. I was Zero Cool." "Zero Cool crashed 1.507 systems in one day, biggest crash in history. Front page New York Times August 10, 1988"
Never forget today, when, on this day in 1988, Zero Cool crashed 1,507 systems in one day.
10.08.2025 15:28 โ ๐ 1007 ๐ 378 ๐ฌ 9 ๐ 47For once, a very good headline, and of course @paolobacigalupi.bsky.social is not wrong here
www.cpr.org/2025/07/12/i...
So yesterday on X someone from X engineering tweeted at me that X does, in fact, use HSMs and the key ceremonies are โcoming soon.โ Iโve updated the post but Iโll be honest this whole thing doesnโt fill me with good feelings.
10.06.2025 13:29 โ ๐ 21 ๐ 1 ๐ฌ 5 ๐ 0Regardless of how good or bad their Juicebox deployment is, at the end of the day, the client code has access to the unencrypted text and/or private key and can do whatever it wants with it.
10.06.2025 15:39 โ ๐ 0 ๐ 0 ๐ฌ 1 ๐ 0And as you mention without an independently verified key ceremony, there's no way to know if the realm is running on commodity hardware, a poorly configured HSM that can leak keys, or a correctly configured HSM.
09.06.2025 21:27 โ ๐ 3 ๐ 0 ๐ฌ 1 ๐ 0I wrote a bit more about Xโs new encrypted DMs and the Juicebox protocol. blog.cryptographyengineering.com/2025/06/09/a...
09.06.2025 18:46 โ ๐ 71 ๐ 27 ๐ฌ 6 ๐ 0Juicebox had 2 realms running on real entrust HSMs managing billions of (test) keys. The impl is complete. That said Iโm not aware of any deployments of it outside the ones Juicebox ran.
09.06.2025 20:00 โ ๐ 5 ๐ 1 ๐ฌ 1 ๐ 0