Hacker News

Yeah. I just dropped another repackaging repo for Wispr Flow. https://github.com/wispr-flow-linux/wispr-flow-linux

A lot of that is keyboard shortcuts for push-to-talk. Easy right?

X11 is mostly fine, but the world is moving into Wayland. Wayland doesn't have shortcuts native and relies on xdg-desktop-portal, which in turn relies on each backend to implement it's own version.

COSMIC from the Pop!OS team's xdg-desktop-cosmic doesn't support GlobalShortcuts yet (might now, haven't checked in a bit). So XWayland for them.

Tray icons? GNOME doesn't have a tray out of the box, but there's an extension. There's no standard for whether it's light mode or dark mode across distros and when you map out the options, no api's indicate whether the tray is light or dark while in light/dark mode. At some point you have to just accept it's not always perfect or patch in an override.

Reminds me that I occasionally have to set _JAVA_AWT_WM_NONREPARENTING=1 because it's not always inherited from my login shell. Otherwise Java windows won't display anything because I suppose Java waits for them to be reparented.

> they’re running an unknown distribution on a 13 year old ThinkPad.

"Tony Stark can do it in a cave! With rocks!"

This does feel like the perfect setup for Claude though.

Much easier to create a vm testing swarm of 100 disitributions with llms

Did you hire a Linux release engineer? Or was the situation the typical team of devs maining macOS that have never heard the term “Wayland” before plus That One Guy who switched to Ubuntu last year and advocated for it?

There are companies that do this right. But it often requires a hire. Too many companies think they can just yolo it because Linux isn’t a serious OS or whatever and then are surprised when it doesn’t work out well.

You compile for the lowest possible Linux kernel and bundle your libs. Don't use container formats for stuff like this. tar.gz with an installer script is king.

I dunno why this is always so difficult.

[flagged]

Flatpak mostly solves this for GUI apps.

Electron apps don’t work well across all of the Linux distributions if you’re doing anything that isn’t very simple.

The comment was that the Electron apps aren’t being released for Linux even when they exist because Linux is so much harder to support, even in Electron.

If they don’t have resources (or desire) to keep the Electron app working on all the Linux distros then they definitely won’t have the resources to write a completely separate GTK app for the few Linux users.

We have basically achieved AGI, but typing on mobile is still an unsolved problem. GBoard's dictionaries for Czech and Polish are still missing many usual declensions...

The old minuum keyboard was fantastic; now I'm forced to use a swipe keyboard I'm constantly making mistakes - but at least its faster than pecking.

For future reference: you can edit posts for up to 2 hours.

Nice, you have RPMs and DEBs in a remote repo we can add! Thanks for making it so easy to use :-)

Also, I can't break the swipe keyboard habit either. It's the worst, but still better than the alternatives. Someday I hope physical keyboard makes a return (but I"m not holding my breath)

I've looked on the rare occasion, but no one is asking for it on the repo, so hadn't been a priority like other distribution channels.

It's great that I can ship one item for all platforms, but Flatpack doesn't solve the compatibility discovery problem for me.

More context in my reply to the comment linked here :https://news.ycombinator.com/item?id=48434436#48435661

Everyone parrots this but I don't think it's true. The Linux kernel does famously have a stable ABI (we "don't break user space" after all).

The issue is folks expect this to be at a higher level, so when libc or GTK or Qt etc. have breaking changes, all your apps using the old versions fail. This is a legitimate pain point with traditional distros.. I don't want to sound like I'm downplaying it.

However, this is basically solved by flatpak (and others like it, eg snap) which contain _all_ these dependencies in the package. Layering (ala containers) is used for deduplication so you don't have 20 copies of a given GTK version.

While MacOS provides the windowing toolkit etc. at the OS level, it's otherwise similar to how a .app file works. Installers aren't dropping dynamic libraries and resource files all over your disk, the app is "self-contained."

This isn’t an issue in practice because the software running on Linux is open source. Yes if you want to distribute proprietary binary blobs and have them work forever it’s going to be a challenge, but in that case better to stick with the binary blob operating system.

The stable Linux ABI is Win32 provided by Wine.

Anthropic claims 8-fold productivity increase since 2025. If even that isn't enough to enable support for Linux, I don't know what it is.

If only Anthropic had some kind of automated testing, support, planning machine.

Those highly paid human devs are hard at work on the Torment Nexus, which is their priority of course, you don't want China to win do you?

You’re asking too much.

> All that said - I actually use the CLI for nearly everything (even on Windows).

I also haven't touched routines, but I use cc to write automation tasks that will integrate a model when I need an inference layer. Which I also did before routines..

Have people actually been using routines effectively?

> I don't think the CLI offers daily routines under the Anthropic subscription anymore?

It (Claude Code) does, I discovered it by accident recently, having never used daily routines before. Haven't touched Claude Desktop at all, outside of playing with it for 30 mins or so months ago.

TLDR: I used Claude Code to build a command that scrapes job postings from a few employers I am interested in (it is a bit more complicated than that, but that's the gist). At the end CC asked me "do you want me to re-run it daily?" I said yes, and it generated a daily routine and gave me a URL to my anthropic account page where I can see all my daily routines.

There, it says that I am currently using up 1 out of 15 "free" daily routines that come with my personal subscription, and I would have to pay extra if I want to have more than 15 active at a time (I assume by switching to per-token pricing for anything beyond 15, but not sure).

> There's also the cross conversation memory search, which uses a different conversation dataset (the Claude Web / Claude.AI conversations) than Claude Code does. I'm not even sure Claude Code does cross conversation search?

This is one of the first things I “fixed” with skills and hooks. I index every conversation in SQLite and have a skill which knows what to do when I ask it to search the index. I had to avoid the word memory because it’s too tied up in other parts of the context. It even indexes across my different machines. I set this up because I have terrible context discipline. I’ll go off on a tangent in one context and start planning and sometimes implementing something based on that thread which really deserves its own context. Afterward I can create the new context and move relevant bits to it, but I’d lose that initial starting conversation which inherently has more data than the summary in the new context.

I also use a few different related contexts. One where I’m building a game engine in zig and another talking about game ideas. There’s a lot of back and forth going on there which needs some shared context. I solve this with a combination of Claude.md references and that searchable session index.

Everything I do with scheduled tasks are just wired up with systemd and simple scripts. No LLM in the critical execution path. Again a skill tells CC how I manage those scheduled things so I just have to say something like “run this every day at midnight” and CC has reliably taken care of the rest.

> Scheduled tasks that run locally ( https://support.claude.com/en/articles/13854387-schedule-rec... ) importantly different from Claude Code routines

What do people do with these? I don't use Claude but when I did I couldn't think of anything useful to do with the routines. I'm probably not being imaginative enough.

There are a number of utilities for this. I use jai: https://jai.scs.stanford.edu/ but also have seen nono: https://github.com/always-further/nono smolvm: https://github.com/smol-machines/smolvm zerobox https://github.com/afshinm/zerobox and matchlock https://github.com/jingkaihe/matchlock

They all have pros and cons. Pick the one that suits you best. Then you're also agent harness flexible (I use opencode).

The cli works on regular sandboxes just fine (podman, docker, bwrap, etc).

Sandboxing a GUI is typically more operational overhead than sandboxing a cli (mounting compositor sockets, GPU access, etc).

I made myself a very simple one from the start when I realized it can access everything on my computer https://git.jeena.net/jeena/agent-container my goal was that it would work transparently and the paths and user, etc. would be just the same as on the host but inside of a docker container.

I've been using jai [1] for sandboxing on linux (although I use opencode and local models and not claude code) and I'm pretty satisfied with it. It comes in three different modes [2]: casual mode, strict mode, and bare mode. Here's some descriptions of each mode:

Casual mode [3]: > Your home directory is mounted as a copy-on-write overlay. The jailed process sees your real files, but writes go to $HOME/.jai/default.changes instead of modifying originals, except in the directory where you ran jai. Your current working directory grants full read/write access to code in the jail (unless suppressed with -D). So files deleted there are really gone. /tmp and /var/tmp are private. The rest of the filesystem is read-only.

Strict mode [4]: > The process runs as the unprivileged jai system user, not as you. Home directory is an empty private directory at $HOME/.jai/<name>.home. Granted directories (via -d or cwd) are exposed with id-mapped mounts — files look like they are owned by jai inside the jail. Because the process has a different UID, it cannot read files outside your home directory that are only accessible to your user — this is where confidentiality comes from.

Bare mode [5]: > Home directory is an empty private directory, like strict mode. But the process runs as your user, not as jai. This means it cannot provide confidentiality — the process can still read any file accessible to your UID outside the home directory.

I've always ran my stuff in casual so far just so my whole computer doesn't get rimraffed :P. but I'm thinking of switching to just strict mode, but haven't really vibe coded in a while so I haven't tried it yet.

[1]: https://jai.scs.stanford.edu/

[2]: https://jai.scs.stanford.edu/modes.html

[3]: https://jai.scs.stanford.edu/modes.html#casual-mode

[4]: https://jai.scs.stanford.edu/modes.html#strict-mode

[5]: https://jai.scs.stanford.edu/modes.html#bare-mode

If you don’t trust the CLI version to be properly sandbox d, why would the desktop one be?

does claude desktop actually solve this issue? I’m on mac and use docker sbx to solve this https://docs.docker.com/ai/sandboxes/get-started/

On Linux you have bubblewrap!

Some more details*

It it was an LLM, why didn't it just vibecode the desired software instead of generating ticket spam?

In the way of eschewing all the APIs that allow an X11 client to control your desktop.

pull requests are welcome