Hacker News Reader

Comment by Lendal | Hacker News Reader

Article did not cover the "Why" question. WSL solves the problem of using Windows while still having a fast Linux/Docker development environment always at your fingertips. I'd understand the "because it's there" argument, which is a really great reason to do just about anything. But is there any other reason why you would need to use a Linux GUI on top of Windows?

View on HN

Because if there is something that you can do from an actual Linux install (VM or otherwise), and not WSL, then WSL is not a complete solution, and most devs will see it as such, whether they have a pressing need to run gui apps or not. If it can't, then it's not real.

Definitely. I was totally put off by wsl1 when I could apt install foo, but foo wouldn’t run because shim X was missing. For one, no haskell based apps was working in the beginning, full stop. I also had problems with git-lfs losing files somehow. I stopped using it as fast as I could hit ctrl-D.

Personally, it's because Linux is a wonderful development environment that meets my needs better than windows. Plus emacs (and most linux-first editors, for that matter) is lightyears faster on wsl than on the native windows builds.

The only issue is with WSL I'm limited to console apps, so if I want to build an OpenGL or Vulkan or GUI project, I have to use Visual Studio or VSCode if I want a good development experience, and even then, it's not quite as comfy.

I can only hope that with better GUI support comes some kind of GPU passthrough, but a woman can only hope.

I can think of cross-building. You could for example build a QT app and test it directly on Linux from your main Windows machine.

Reply on news.ycombinator.com