Hacker News new | ask | show | jobs
by tastysandwich 1040 days ago
A lot of Ubuntu hate, and I totally understand why. We all remember Mir and Upstart. And now Snap. So if this bothers you, by all means look elsewhere.

On the other hand... I've used Ubuntu desktop consistently for about 15 years now. I'm not a power user by any stretch. I just want things to work, and the "just works" factor has improved considerably.

Installation is ridiculously easy. It's fast. It never crashes. It finds all the right drivers for me. Suspend works. Audio works great, including with recording software like Reaper/Audacity.

I just don't have time to mess around with stuff anymore. Gone are my Arch Linux days. I get about half an hour of time after putting our little one to bed, and I want to spend 0% of that on configuring stuff.

2 comments

I have the same experience running good old fashioned Debian.

It just worked out of the box on both my desktop machine and Framework laptop. Detected all the necessary drivers meanwhile wifi, touchpad, and all the usual suspects just worked.

I've been doing most of my work on Debian for a pretty long time, even using Debian in WSL2 when I have to use windows, and I can't recall having met with a single problem that using Ubuntu would have solved.

> We all remember Mir and Upstart

What was wrong with Upstart? It was great, it improved boot times and everyone adopted it, even RHEL.

It was then superceded but it was a great improvement over SysV Init.

Yeah I actually appreciated how easy it was to create daemons with Upstart. I had no problem with it, I just remember a lot of people pissed off that Canonical wrote their own init service instead of using systemd (which they eventually moved to, and which I recognise people also have very strong feelings about).
Upstart came along about 5 years before the initial release of Systemd, so I don't think your memory is correct.

Systemd came as an evolution of Upstart, but Lennart and Kay claimed that there were design limitations to Upstart that prevented it from going as far as they want.

I'm not going into whether that was true, or just NIH, or whether Systemd is going in the right direction.