Is there really such a thing as a failed experiment? Don’t you just find a closed door at the end of it and isn’t knowing the door is closed worth something?
That’s like what got me started with Linux. I was a Windows guy at college and Vista was going to update what I loved about XP. So I bought a machine with the ‘designed for Windows Vista’ sticker and then suffered through a tremendously frustrating experience that was alleviated by rediscovering Linux. For me the Linux experience has almost always been via the desktop, not the server application most folks know it for.
Two weeks ago I decided to try and use the Windows’ enabled versions of Linux via Hyper-V and WSL. This was in part inspired by Jason’s challenges on Forbes and the podcast ChooseLinux (although in reverse). On Saturday I partitioned my hard drive and reinstalled KDE Neon so I could dual boot again. I made it two weeks and two weeks was too painful.
As you’ll remember from my first post I’m a casual Linux user. It solves problems for me in the home, it’s fun to tinker with, but I’ve never written code and my shop at work is mostly Windows. Most of my output at work is done using MS Office.
There are likely very good reasons for code writers and seasoned professionals to have Ubuntu so easily installed via Hyper-V and the WSL ready to go. In some situations I could imagine that having these environments will allow people to mock-up and test applications on the same machine that has their MS Office install. But in that scenario they’ll still probably need another Linux machine to actually get the work done.
Also, I’m not one of those guys.
I love the networking tools Linux has. The Ubuntu Podcast taught me
sudo netdiscover which shows the machines one one’s network. Similarly Kali Linux available via WSL has lots of fun networking tools, but none of these networking tools work in the WSL because it doesn’t have access to the network stack–so for my purposes it makes it virtually useless.
While the networking tools do theoretically work in Hyper-V they give no practical output because the Hyper-V instance isn’t bridged (like you could set if you were running Linux in a VirtualBox instance).
What I do appreciate is that Microsoft was willing to invest in this first historic iteration of making parts of a very powerful operating system available on its platform. When it comes down to it the parts they’ve enabled though, aren’t the parts I value.
I can’t say that I’ve conclusively proved that Microsoft’s implementation of Linux under Windows can’t work for some people. My use case is narrow. The functionality provided by Microsoft’s implementation of Linux is also narrow and there’s little overlap between the two.
I think this is just another example of what makes Linux great. It satisfies so many broad use cases! Microsoft’s implementation can’t possibly take over the breadth or even a majority of the cases out there where Linux solves real problems for people.
My use case is mostly outside of their current offer, but I’m grateful they’re offering what they do. I now have two applications I can run on my Windows machine that I didn’t have before: SSH and youtube-dl.
SSH works as expected.
Youtube-dl is more emblematic of how I see the whole experiment. It functional, but not elegant. Once a file downloads from youtube-dl all I have to do is navigate to
C:\Users\username\AppData\Local\Packages\CanonicalGroupLimited.Ubuntu18.04onWindows_79rhkp1fndgsc\LocalState\rootfs. Which is so intuitive right?
Maybe there’s more functionality buried in what they’ve provided, but I don’t think the time it will take to figure it out is as effective as rebooting into KDE Neon.