• 2 Posts
  • 62 Comments
Joined 3 years ago
cake
Cake day: July 6th, 2023

help-circle




  • It depends on the subject area and your workflow. I am not an AI fanboy by any stretch of the imagination, but I have found the chatbot interface to be a better substitute for the “search for how to do X with library/language Y” loop. Even though it’s wrong a lot, it gives me a better starting place faster than reading through years-old SO posts. Being able to talk to your search interface is great.

    The agentic stuff is also really good when the subject is something that has been done a million times over. Most web UI areas are so well trodden that JS devs have already invented a thousand frameworks to do it. I’m not a UI dev, so being able to give the agent a prompt like, “make a configuration UI with a sidebar that uses the graphql API specified here” is quite nice.

    AI is trash at anything it hasn’t been trained on in my experience though. Do anything niche or domain-specific, and it feels like flipping a coin with a bash script. It just throws shit at the wall and runs tests until the tests pass (or it sneakily changes the tests because the error stacktrace repeatedly indicates the same test line as the problem).





  • The further we get from 9/11, the more impossible it is to even conceptualize the future. Like, idk what kind of jacked up shit happened to the timeline that day, but it was bad. Maybe tower 7 was the fucking gateway to the multiverse and some Doctor Who shenanigans happened when it fell.

    We’ve reached the point where MTG and Candace Owens are actually saying reasonable things that no one in leadership positions is allowed to say. I sincerely do not know what could possibly happen next.


  • I don’t think you can advocate for anything even remotely on the “right” in political discussions anymore unless you mean MAGA. That well is so poisoned at this point that everyone is going to assume you’re a MAGA troll wearing a mask the second you voice any right-leaning opinion.

    It’s pretty unfortunate. There are plenty of “live and let live” types in the US that identify informally as libertarians and would make great allies.





  • Listed salaries are almost always what the employee pays, not what it costs the company. In the US, this includes the payroll tax, and cost of “benefits,” like healthcare and unemployment insurance, and is referred to as the burdened rate. This is separate from the income tax the employee has to pay to the government, mind you.

    The burdened rate for most employees at the companies I’ve worked for in the US is like 20-50% higher than the salary paid. Not sure exactly how it works in France, but I do know there’s a pretty complex payroll tax companies have to pay. I think it’s something like 40% at the salary you quoted.




  • I’m not sure if you know this, but…that doesn’t fix most of the security issues in the linked list. All the reverse proxy does is handle hostname resolution and TLS termination (if you are using TLS). If the application being proxies still has an unauthenticated API, anyone can access it. If there’s an RCE vulnerability in any of them, you might get hacked.

    I run Jellyfin publicly, but I do it behind a separate, locked-down reverse proxy (e.g., it explicitly hangs up any request for a Host header other than Jellyfin’s), in a kubernetes cluster, and I keep its pod isolated in its own namespace with restricted access to everything local except to my library via read-only NFS volumes hosted on a separate TrueNAS box. If there is any hack, all they get access to is a container that can read my media files. Even that kind of bothers me, honestly.

    The overwhelming majority of Jellyfin users do not take precautions like this and are likely pretty vulnerable. Plex has a security team to address vulnerabilities when they happen, so those users would likely be a lot safer. I appreciate the love for FOSS on Lemmy, but it is scary how little most folks here acknowledge the tradeoffs they are making.