Avatar

DESTROY EVERY COMPUTER

@piratesexmachine420

;; This buffer is for text that is not saved, and for Lisp evaluation. ;; To create a file, visit it with C-x C-f and enter text in its buffer.

I'm going to sound like a total killjoy lmao. Dgmw the "net zero information post" meme was funny the first handful of times I saw it, but nowadays seeing someone pull out that phrase on a post just leaves me thinking like. Well, it doesn't *have* to be a net zero information post.

You can actually learn a lot from it if you're willing to interrogate things such as "why did I completely believe this until I saw someone directly on the thread debunk it?" and perhaps "are there any particular preconceptions the original claim is playing into to Confirmation Bias its way into being believed with no evidence?" and "How susceptible am I to believe something because it confirms biases I hold or it makes for a satisfying narrative?". And then trying to apply the things you learned from that exercise the next time you catch yourself believing a similar claim without looking into it.

I feel like these days at least where I live we take for granted that there’s a wide variety of first names people can have but when you start digging into history you’re like oh with no exaggeration statistically most men in this time and place were probably named John

I’ve talked about this before but like almost every man in Ancient Rome was named either Marcus or Gaius. Possibly even both at the same time. They invented a whole secret third name so they could tell people apart because everyone was named Marcus or Gaius.

Even Cicero was named Marcus, that old bastard.

Also yeah you start looking a little into Arab and Muslim history like I have recently and every other guy is named Mohammed. It’s still the most popular given name in the world. If you need to guess a random human man’s name and you know literally nothing about him, guess that his name is Mohammed.

The most popular surname is Wang. This is a great example of dependent variables. There are 150M people with the given name Mohammed and 100M people with the surname Wang, but there are only a couple thousand people named Mohammed Wang

Perhaps this is just the "Muhammad Wang" fallacy but I feel like a lot of the people I see complaining about Rust's unsandboxed build scripts are also the people advocating for more hand-written makefiles.

I just don't understand how these people don't realize that nontrivial build orchestration without arbitrary code execution isn't possible. Why are you building any code you believe might be malicious outside of a VM or container? Are you stupid?

...do you know how compilers work?

Saw someone on HN claim "the JVM ecosystem" doesn't have this problem.

Like, sure, if you only link to binary dependencies, they can't run code before link time -- but you have to reframe this issue in your head as "some of my dependencies ship tooling" and not "in their hubris some of my dependencies are trying to Seize Godhead".

Recognize the contradiction when you trust CC "just because", but not build.sh "just because", and reconcile these positions; either by sandboxing every build and all software, by researching your deps and making an informed decision about their trustworthiness, or by admitting when you don't care. (Or realistically some combination of the three.)

Perhaps this is just the "Muhammad Wang" fallacy but I feel like a lot of the people I see complaining about Rust's unsandboxed build scripts are also the people advocating for more hand-written makefiles.

I just don't understand how these people don't realize that nontrivial build orchestration without arbitrary code execution isn't possible. Why are you building any code you believe might be malicious outside of a VM or container? Are you stupid?

...do you know how compilers work?

Perhaps this is just the "Muhammad Wang" fallacy but I feel like a lot of the people I see complaining about Rust's unsandboxed build scripts are also the people advocating for more hand-written makefiles.

A vote for me is a vote to bomb the realm of forms with such brutality that the traumatic memory outlasts the sun and the stars. Generations will live and die until the years won't even have names and even then up against the ragged right edge of history the word "meaning" will be enough to send grown adults into hysterics.

Notes on Racket so far:

  • It's gonna be a while before I get used to these square brackets.
  • Lisp-1s are undeniably more elegant, but this still feels weeeeird.
  • The lack of nil keeps tripping me up, as does the quoting behavior of null. In CL or Elisp, 'nil eqs nil eqs '(). In Racket null eq?s '(), but 'null doesn't, like a normal symbol. Hrrrm.
  • '() being truthy instead of falsy is also fucking with me.