This is extremely cool. The basic idea: WireGuard is just a network protocol, like any other, and you can drive it from unprivileged userland code… which means you can drive all of TCP/IP from unprivileged userland code, through WireGuard.
Why would you ever want to do that? Well, we expose services on Fly.io over WireGuard (and, for security, over no other interfaces) but not all of our users are going to install OS WireGuard.
But: all of our users have our (Golang) `flyctl` installed, and flyctl can do WireGuard via wireguard-go, and then userland TCP/IP, to be a client of a network service exposed over WireGuard, without installing WireGuard itself.
Obviously, you should just install WireGuard. But for the stuff I’m talking about, perf doesn’t really matter (if it does, again, install WireGuard), and convenience really matters a lot.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Illinois makes it super easy to send FOIA requests to any municipality (just look up their FOIA officer’s email); it’s free, and they get just 5 days to respond (10 with a written extension) before you can sue and have them pay your legal costs if you win.
What I’m saying is, not a crazy project to just come up with every police officer in all of Chicagoland who took PTO during the riots in DC.
This paper is very cool: behavior oracles in interactive systems that reveal successful decryption can, with a bunch of different AEADs incl. GCM and Chapoly, discern which specific key was used in something resembling log k queries. eprint.iacr.org/2020/1491.pdf
It’s based in part on the idea of “non-committing AEADs”, which are, roughly, AEADs where the specific key used to encrypt isn’t encoded into the output. For something like GCM, this means it’s straightforward to generate K_1, K_2, and C which decrypts under K_1 and K_2.
I found Shay Gueron’s writeup on key committing AEADs to be pretty accessible (I’m just reading casually), with worked examples. eprint.iacr.org/2020/1153.pdf
Mudge is the new head of security at Twitter, which got me talking about cDc, hacking groups, cliques, and the distinctions between them. I mentioned 8lgm and TESO as examples of hacking groups best understood as hacking groups, unlike cDc.
Someone said: “never heard of them”.
This creates an opportunity for me to talk again about my favorite exploit of all time, unquestionably a part of the canon of our field.
The year is 1995 and BSD Unix runs the Internet. The most important hacking target is SunOS 4.1.3; every network you want to get on is running it somewhere, and often everywhere.
The most important SunOS security research group: 8lgm.
Kind of crazy watching the orange site, which believes I’m an NSA stooge, fall over itself arguing that publishing DKIM keys to provide deniable email would be a grave injustice, depriving “activists and historians”.
This is what happens when you have a culture that attempts to derive everything axiomatically, just moments after reading something. They forget that deniable messages are literally part of the premise of messaging cryptography. otr.cypherpunks.ca/otr-wpes.pdf
This is currently the top comment on the thread. Again: these people think I’m a shill for NSA.