deleted by creator

the OS has included a path_helper utility. This reads entries from /etc/paths as well as from any files in the /etc/paths.d directory. While one could potentially take advantage of this system, I (and I’d guess most others) handle adding new PATH entries using the .zshrc file, or similar

I tried to use path_helper once, because it seems like a good system. The issue that I realized upon attempting to edit these files is that they are root-owned system-level files. User-level path management needs to continue using shell files (.zshenv in my setup).

score 1 · canonical

I don’t love this change but I think it’s an improvement and I really appreciate the work that goes into moderating the site. Thanks.

score 1 · canonical

The tag is about a year old and on the post introducing it people pointed out that it was inaccurate. The tag is supposed to be a neutral tag to be used for any discussion of using coding agents, but the word vibecoding has a negative connotation. It seems like renaming (to ai-usage) or redefining the tag (to “vibe-based LLM use”) would be very reasonable.

Some people in this thread argue that all posts about LLM usage should have a tag with a negative connotation attached, because LLM usage is negative. The issue with this line of thinking is that, as things stand, they are attempting to leverage the confusion between the intended and actual usage and meaning of the tag. For example, if person A tags a story with “vibecoding” as a neutral descriptor, and the author of the story (B) takes offense, because of “vibecoding”’s negative connotation, person C might say “well, the author’s LLM use was bad. Who cares that A didn’t mean it that way, I got the outcome I want.” The issue with this dynamic isn’t that B is offended. The issue is that C is arguing at the expense of A. Having a tag with a negative connotation isn’t bad because it hurts AI user’s feelings, it’s bad because it’s impossible for someone with a neutral or nuanced opinion on AI to accurately tag things without appearing to support a more vitriolic position. Rather than hurting LLM users, this hurts the community as a whole by leading to endless arguments about who should and shouldn’t be given the thorny LLM tag.

I’m actually optimistic about this moving forward. Because most people, AI-proponents and haters want the same stuff on the front page. We all want good posts about making good software. The hallmark of “high conflict” or over-polarized groups is that they’re willing to hurt their own goals in order to deliver (what is perceived as) a worse outcome for their opponent. (i.e. “if I can’t have it then no one can.”) We’re not there, thankfully. Even people with this thread with extreme anti-LLM views are not trying to intentionally hurt the community.

score 3 · canonical

“Install a custom kernel and short the reset pin to ground” feels pretty sophisticated for an anti-cheat bypass.

Is it possible to package all of this into a virtual machine that emulates the relevant hardware?

score 1 · canonical

I don’t understand this person’s threat model at all. Why do they trust the person that they’re emailing but not an email provider?

score 1 · canonical

My current programming project is an implementation of the beautifully complex chess-like Veney.

score 1 · canonical