Hacker News new | ask | show | jobs
by Ygg2 394 days ago
Linux biggest problem as a consumer OS was rarely tech and mostly just the schizoid amount of options, and lack of consensus on what to use.

In essence Linux suffers from a Lisp curse. Whenever two OS nerds disagreed on something they made their own slightly different distro.

This means wasted effort on multiple DE, window managers, app flavors, installed libraries. To this day, almost no two distros can agree on baseline libraries every Linux must have.

4 comments

> Linux biggest problem as a consumer OS was rarely tech and mostly just the schizoid amount of options, and lack of consensus on what to use.

It's only a problem if you think it is. In practice I use at least 3 or 4 different distros on a daily basis and I never have any issue juggling between them. For most of the typical use cases it does not even matter, and on the desktop side flatpak resolves many issues.

> It's only a problem if you think it is.

No. It's a problem, if you as distro maker support non-technical consumers as well.

Imagine troubleshooting Windows but you also have to figure out which DE, WM, libraries the user updated and so on.

> desktop side flatpak resolves many issues

You mean AppImage, Snap, etc.

The point of a distro is to select what you support. SteamOS only has one DE etc.
Except the user-base wants infinite customizability, and users are often to mess with it. And getting the code to just work together nicely is a nightmare, where OS updates can break drivers, forcing you to try to jerry rig a solution that partially works.
You let something like Arch target those users that want infinite customizability. SteamOS certainly does not offer that.

And if there was such a nightmare to create a distro, how come there are so many?

I love how in this context people call it wasted effort while in other areas is just competition.
Competition works because the more successful a company is, the more resources it gets (money from the customers).

If a free Linux distribution is more successful, its resources don't scale accordingly.

Yeah, duplicating/triplicating/n-cating bugs, feature development and support effort, really paid off for the Linux desktop ecosystem. Which year will be the year of the Linux desktop? One, when we get brain to brain communication and abandon desktops entirely?
Cooperation/Symbiosis: win-win

Predation: win-lose

Competition: lose-lose

This is how dynamically coupled systems work.

Linux is not a full OS. It is "only" an OS kernel. Linux can't replace Windows. Fedora or Ubuntu can.
You missed GNU!
I don't get your point.
Haha, you may be too young I'm sorry; it was meant to be a joke.

Back in the 90s when linux came out and started getting traction, Richard Stallman was adamant that people should call the operating system GNU/Linux , because Linux was only the kernel, but mlst of the userland utilities were the GNU software (which were planning to make a full OS like GNU/Hurd).

Some people made fun of that, and I was just kind of paying the joke.

The biggest problem of Linux as a consumer OS is that Linux is not even an OS, it’s just a kernel.