<pie_>
if anyone wants to drop in im going to try to chill and read some textbook and maybe fall asleep <ehmry> pie_: mumble.nix-community.org is still up
ghavil has quit [Quit: Ttyl]
ghavil has joined #spectrum
cole-h has quit [Ping timeout: 240 seconds]
<IdleBot_6d92ac96>
Coooool, I can create a process that is not killed by Alt-SysRq-k at VT3 but is killed by fuser -k /dev/tty3. My assumptions are broken once again…
<IdleBot_6d92ac96>
Re: phones: now that the sound seems mysteriously fixed, I should also retry to boot Debian on Cosmo, not just live in it inside Termux and chroot…
<pie_>
i dunno if yallve seen any prior work on this (i havent but im 80% certain it exists) but like, what do you think about a new abi between processes that is more typed and can do more than stringly typed stdio and taking stringly typed params and returning an int
<pie_>
ive got some vague ideas floating around in my head but i realize im pretty low on the "good at systems stuff" ladder
<pie_>
how dumb of an idea is this? shitty hsm via something like keepass on a usbarmory controlled via vnc or something with some other back channel into the machine to type passwords and whatnot
<pie_>
construction. In particular, he introduced the often-cited separation-of-concerns design principle, which, he remarked, “even if not perfectly possible, is yet the only available technique for effective ordering of one’s thoughts, that I know of.”74 Following the example of Hoare and Wirth, he also advocated for various forms of abstraction and the use of assertions to annotate programs. At a later stage, he argued that the programming
<pie_>
https://inference-review.com/article/the-man-who-carried-computer-science-on-his-shoulders#footnote-72 In the years that followed, he produced a number of engaging and influential essays on software development in which he explicitly referred to the software crisis as an urgent problem. He forcefully argued that software systems should be built on sound design principles, and that correctness should be a driving principle behind program
<pie_>
process itself should be viewed as a mathematical activity.
<pie_>
Dijkstra said it; use compartmentalization kids :P
<pie_>
unrelated, someone told me to use https://eprint.iacr.org/2018/695.pdf SPHINX instead of whatever crappy thing i can concoct for compartmentalizing my password manager
<pie_>
havent read through the article yet, i really hope its something actually usabel
<pie_>
*as opposed to trying to run my keepass on a raspi or something
<zgrep>
I... don't understand how anything can be resistant to a keylogger. Is that one of the things they're claiming, or am I misreading it?
<pie_>
dunno, probably
<pie_>
im preoccupied so i havent really looked to figure out wtf is going o
<pie_>
*on
<zgrep>
Oh, okay. It's resistant to an attacker having unfettered access to the keystore (which can be an online server, or local), but not to intercepting/modifying the application running on the end-user machine / the application that gets the user password.
<zgrep>
So it's safe against a person-in-the-middle attack between the user and the keystore, as well as attacks on the keystore, and offline (not during use) attacks on the user.
<pie_>
probably easier for me too
<pie_>
well, papers are supposed to be dense so...
<zgrep>
Yeah, but I should get better at... searching through them for the key info. :P
<pie_>
hm
<zgrep>
Hm. Does the IRC logger have a no-log-for-this-message keyword?