Thomas H. Ptacek
@tqbf
Followers
33K
Following
5K
Media
2K
Statuses
92K
Don't look at me sideways. Don't even look me straight on. bsky:@sockpuppet.org
✶ ✶ ✶ ✶
Joined October 2007
DOWNVOTES IMPRISONING ME ALL THAT I SEE ABSOLUTE HORROR I CANNOT LIVE I CANNOT DIE TRAPPED IN THIS THREAD COMMENTS MY HOLDING CELL
13
49
523
Not super worried about implications of LLMs on work sample hiring processes; you can build work samples that assume LLMs. I am very worried about the impact on Slack/chat interviews, which are more humane than voice interviews but very hard to LLM-proof.
1
0
3
This is literally the worst thing I’ve ever seen in my life
266
3K
84K
My attempts to explain dynamic IP addressing to Joyce Carol Oates have not gone well.
@JoyceCarolOates No, really. Almost all consumer Internet is via dynamic IP address, which means that the computer on your desk can be given a new Internet address every time it connects to the network. https://t.co/6dVTE5iBqx
121
161
5K
Final SCW of 2025! We had Matt Bernhard on to talk about cryptographic voting systems, in the wake of the IACR election. (Everybody I voted for in the new election won! Woo!) https://t.co/TxqYFCc5DG
securitycryptographywhatever.com
The International Association of Cryptologic Research (IACR) held their regular election using secure voting software called Helios…and lost the keys to decr...
0
2
13
Whoah, whoah, whoah. Everybody, slow down. Bernie Sanders hasn't been consulted about your new product! We need to check with the East Coast Longshoremen's Union before we decide if it's OK to bring it to market.
Bernie Sanders: "I think we need to be thinking seriously about a moratorium on these data centers. Frankly, I think you gotta slow this process down. It's not good enough for the oligarchs to tell us, 'It's coming, you adapt.' What are they talking about? Are they gonna
0
2
22
Cryptographers don't trust state machines, demanding formal verification even for protocols designed to be simple and legible. GPG has state machines even its authors don't understand, and so you can trick it into verifying unsigned messages.
6
28
222
The boy's croissants succeeded. He's a biophysics grad student and a running theme of the last 8 years has been "he's a better cook than me for reasons I attribute to the bench work training I paid for him to get."
0
0
8
@ZacMcCrary Incorrect. The film is a classic regardless of who reruns it. You’ll tweet you eye out kid.
2
5
151
Rust people should be embarrassed by the people trying to argue that Rust is the only truly memory-safe mainstream language. Far and away the cringiest RESF trope.
4
3
22
Qualified success. Next time don't pack them tight in a pan like you would normal cinnamon rolls; spread them out like croissants. They are indeed semilaminated croissinnamon rolls. Butter leakage is cinnamized, went on waffles.
1
0
4
Yeah by the 3rd turn it looked pretty hopeless, but I learned something new, which is that you can take any failed croissant batch and turn them into semilaminated cinnamon rolls.
1
0
2
First batch, early failure! Butter book too thick, so the first roll-out left butter peaking through the seams. We'll see how it goes, but starting my second batch preemptively.
2
0
2
Forget the stupid ideas, I don't get how anybody could sleep after writing a paragraph this badly.
SCOOP: Internal docs obtained by @thecrimson show Harvard Salient leaders casually used racial slurs, circulated a draft with Nazi symbols, and criticized women's voting rights — revelations that triggered the board to suspend the conservative magazine. https://t.co/JoEpoeFGON
2
0
8
This time I'm going to beat the butter in a stand mixer, form it back into a square and briefly chill it, instead of trying to make a butter square with a rolling pin. Convinced this is the source of most of my leakage: too-cold butter.
1
0
2
nooOoOoo the quizzes still have trig inverse integral problems in them death first aahewhhwharhgrg
1
0
3
I am bad at math and so am asking this sincerely: why is cofactor expansion taught? What's the point? I would just reduce the matrix to take its det and so would a CAS.
2
0
6
I want to say again that I love MathAcademy and having finished MF3 I now have more single-variable calc (and more enmity for trig identities) than I ever thought possible. But MML starts with fucking Laplace determinants and I am in a holiday murder fugue.
1
0
12
Everyone who spends a day among the zoomers comes back talking like they glimpsed an eldritch horror at the bottom of the ocean.
9
15
313