I want to run a shell script that might open my browser to a specific website. I don't want the page to load when this happen. But I cannot switch off my internet access also (as I use the internet to remotely access another system at the same time). So I am planning to isolate the run time environment for the shell script.

I an on Arch and I used to use a AUR package called bubblejail to do this. But with the whole AUR security fiasco, I am not trusting any packages from AUR. I can switch to another distro if needed, like Rocky or something.

So my requirement is, Internet sandboxing for a terminal and the processes it spawns. Preferably using flatpak commands.

Edit: I tried disabling the internet usage for a terminal from Flathub using Flatseal. Sure I cannot curl after this, but when I launch my browser using it, it had Internet access.

 

Decided to create a thread for tracking and sharing the news and opinions on the new Malicious Atomic Arch NPM Campaign in which more than 1600 Arch Linux AUR Packages Hijacked to Deploy Infostealer and eBPF Rootkit.

Find the infected packages: https://md.archlinux.org/s/SxbqukK6IA

Most popular packages on the affected list

Package        Popularity                Affected                 Reverted
libgdata           16.98% (2026-06-11 14:59+00:00) (2026-06-11 17:30+00:00)
python-future       5.38% (2026-06-11 15:58+00:00) (2026-06-11 16:54+00:00)
gdl                 3.36% (2026-06-11 13:35+00:00) (2026-06-11 17:32+00:00)
libquvi-scripts     2.31% (2026-06-11 15:05+00:00) (2026-06-11 17:33+00:00)
libquvi             2.22% (2026-06-11 15:04+00:00) (2026-06-11 17:33+00:00)
gtkimageview        2.19% (2026-06-11 13:44+00:00) (2026-06-11 17:33+00:00)
python2-pyparsing   2.02% (2026-06-11 14:23+00:00) (2026-06-11 17:40+00:00)
python2-appdirs     1.96% (2026-06-11 14:22+00:00) (2026-06-11 17:26+00:00)
compiler-rt19       1.95% (2026-06-11 14:23+00:00) (2026-06-11 17:30+00:00)
python2-packaging   1.90% (2026-06-11 14:21+00:00) (2026-06-11 17:38+00:00)
wine-nine           1.86% (2026-06-11 15:48+00:00) (2026-06-11 21:36+00:00)
clang19             1.86% (2026-06-11 15:36+00:00) (2026-06-11 21:24+00:00)
clang15             1.76% (2026-06-12 12:34+00:00) (2026-06-12 12:54+00:00)
mono-addins         1.69% (2026-06-11 15:33+00:00) (2026-06-11 21:34+00:00)
python2-chardet     1.68% (2026-06-12 12:42+00:00) (2026-06-12 14:48+00:00)
python-monotonic    1.55% (2026-06-11 15:43+00:00) (2026-06-11 21:37+00:00)
python2-cffi        1.47% (2026-06-12 12:44+00:00) (2026-06-12 15:10+00:00)
alvr                1.26% (2026-06-11 13:54+00:00) (2026-06-11 16:50+00:00)
python2-gobject     1.23% (2026-06-12 12:44+00:00) (2026-06-12 14:47+00:00)
vidcutter           1.03% (2026-06-11 13:24+00:00) (2026-06-11 17:43+00:00)

Learn more about the attack: https://www.sonatype.com/blog/atomic-arch-npm-campaign-adds-malicious-dependency.

Filen free plan. Any good? (discuss.tchncs.de)
 

I was looking for a Google Drive alternative. Its mainly for storing small documents. 10GB is Filen's limit on their free plan. Its more than enough.

But I am concerned about their privacy. Have anyone used it? I am ready to pay for a really good service but if they are giving it for free than I why should I pay if they are private enough?

They also have paid ones but they are an overkill for me. I mainly use offline HDD backups. These are for some quick access files. I don't need an app or anything. Simple web login would be fine.

 

Is this behavior expected? I would imagine it would be a privacy violation in a multi-user system. I thought they had some sort of encryption for hiding the sites that I visit.

BTW, FF does not do this on private mode. But still it is concerning, that any program can know about the sites I visit just by looking at ~/.mozilla/profile/storage/default.

 

Look, I don't believe that an AGI is possible or atleast within the next few decade. But I was thinking about, if one came to be, how can we differentiate it from a Large Language Model (LLM) that has read every book ever written by humans?

Such an LLM would have the "knowledge" of almost every human emotions, morals, and can even infer from the past if the situations are slightly changed. Also such LLM would be backed by pretty powerful infrastructure, so hallucinations might be eliminated and can handle different context at a single time.

One might say, it also has to have emotions to be considered an AGI and that's a valid one. But an LLM is capable of putting on a facade at-least in a conversation. So we might have to hard time reading if the emotions are genuine or just some texts churned out by some rules and algorithms.

In a pure TEXTUAL context, I feel it would be hard to tell them apart. What are your thoughts on this? BTW this is a shower-thought, so I might be wrong.

view more: next ›