[tptacek]

Yeah, I used to be skeptical of the government provenance of things like Stuxnet (I am not any more, I'm fully sold, like everyone else), and notes like this were why. People used RCS well into the 2000s! RCS as a tool had virtues over SVN and CVS.

[ajju]

My favorite part of the paper is that the “attack” isn’t just exploiting a bug — it’s exploiting how different components interpret the same input. Modifying an executable as it’s loaded into memory is one example, but the deeper pattern is the mismatch.

What’s interesting about the malware in this post is that it goes one step further: instead of exploiting mismatches, it corrupts the computation itself — so every infected system agrees on the same wrong answer!

More broadly: any interpretive mismatch between components creates a failure surface. Sometimes it shows up as a bug, sometimes as an exploit primitive, sometimes as a testing blind spot. You see it everywhere — this paper, IDS vs OS, proxies vs backends, test vs prod, and now LLMs vs “guardrails.”

Fun HN moment for me: as I was about to post this, I noticed a reply from @tptacek himself. His 1998 paper with Newsham (IDS vs OS mismatches) was my first exposure to this idea — and in hindsight it nudged me toward infosec, the Atlanta scene, spam filtering (PG's bayesian stuff) and eventually YC.

https://users.ece.cmu.edu/~adrian/731-sp04/readings/Ptacek-N...

The paper starts with this Einstein quote "Not everything that is counted counts and not everything that counts can be counted", which seems quite apt for the malware analyzed here :)

[Tepix]

Just curious, are you purposely mocking the LLM writing style?

[DANmode]

That’s how everybody in academia, tech, and published authors in general used to write.

Where do you think the LLM is getting it from? ^_^

[red-iron-pine]

the full on em dash requires a different character than - or --

it was generated that way, or else this person happens to know the correct combination of buttons to make that happen.

in 2026, at least 20-40% of social media traffic is bots (and probably higher with better LLMs), so it is usually safer to just assume.

[SAI_Peregrinus]

On Linux, with a compose key, it's <compose><-><-><.> (at least with the settings I have, I don't think I overrode that one). "⸻" is even more fun. You can even make your own sequences, e.g. I've got <compose><O><h><m> for "Ω", and <compose><m><u> for "μ", very handy for electrical stuff like "160μA at 1.8V needs a resistance of 1.25kΩ, dissipating 288μW".

[picture]

In pedantic tradition, I would like to gently remind you that there should be a space between a number and its unit, according to SI standard/NIST.

[macintux]

On a Mac, at least, the "correct combination of buttons" is trivial and easy to remember, even for someone like me who rarely uses em-dash. (But, I want to start using it more because I'm sick to death of people treating it as a scarlet letter.)

[ajju]

Option-shift-hyphen

Thanks for sticking up for my humanity ;)

[IAmBroom]

Microsoft Word changes "--" to an em-dash, by default.

[DANmode]

> or else this person happens to know the correct combination of buttons to make that happen.

Yes, some people care about maximum impact of their words.

> at least 20-40% of social media traffic is bots (and probably higher with better LLMs), so it is usually safer to just assume.

Look around you. If you see 40% bots in HN comments, find a therapist.

[gucci-on-fleek]

> People used RCS well into the 2000s!

I still use RCS today. It's certainly not my preferred option, but my collaborator likes it, and it's not too annoying for me to use.

[NitpickLawyer]

> used to be skeptical of the government provenance

Do you mean skeptical on which government was responsible or that it was in fact a government effort?

I can see how attribution could be debatable (between two main suspects mainly), but are / were there any good arguments against this being a gov effort? I would find it highly unlikely that someone other than a gov could muster up so much domain knowledge, source pristine 0days and be so stealthy at the same time.

[tptacek]

I didn't want to give bumptious government CNE teams that much credit, and also a lot of the indicators people were giving of state origin didn't seem all that predictive. I don't agree with your premise that it takes a state-level adversary to collect the domain knowledge needed to do this stuff, and I certainly don't agree about the "pristine zero days".

[codezero]

I do wonder if these breadcrumbs were also left intentionally. “Oh look, we are using old stuff, don’t be afraid!” Or for some other reason. It is a little surprising to pull off such a sophisticated attack and miss details you could find running ‘strings’ unless I’m missing something and this part was encrypted.

[tptacek]

I think that in the time period we're talking about, RCS wasn't really even all that old. Like, RCS is old, sure, but it was also in common use especially by Unix systems people; it's what you might have reached for by default to version your dotfiles, for instance.

[codezero]

Yes, but even back then I was aware of the sections in executables (wasn’t this where it was found?) and any neckbeard from the 70s and 80s might be even more so aware. That said, yeah, sure, it’s a very possible and understandable oversight, but I’m weary because of all the text in viruses and such as indicators. Seems like a pass over ‘strings’ would be obvious. Though. TIL, strings doesn’t necessarily scan the entire executable.

[breppp]

The same binary has encrypted strings so I assume there was a pass, but if you look at the source control strings they seem to decrease the appearance of maliciousness, even today they are out of place for malware