Linux

You could also ask: "Why are there so many audio formats? Why are there so many video formats?" And so on.

The reason is different people have different ideas on what is the best way to do things.

As another user mentioned, package managers are specific to distributions rather than DEs. The main difference between them is that they're developed by the respective distribution teams, but there are some practical differences too. For example, apt supports versioned dependencies while pacman doesn't because of the different distribution models between Debian and Arch (monolithic vs. rolling release). This affects their dependency resolution strategy with each being better suited for it's respective distribution.

To address your point about package managers being the main difference between distros, this isn't quite true. As mentioned, different distros have different distribution models, priorities, and overall biases/opinions that affect the user experience in a variety of ways and make them better suited to different use cases. I would never dream of putting Arch on one of my servers in the same way that you'd probably never catch me installing Debian on my gaming machine.

xkcd: Standards

Oh no, there are 5 package managers out there and they're all wildly different

I know! I'll make a standard, universal package manager that'll be better than all the others that everyone will use!

There are now 6 different package managers

I wouldn't worry too much about the package manager, just worry about whether the distro has a good package repository. If it has all the software you want to use, then use it. In my opinion, most package managers (dnf, apt, pacman, xbmp) are basically the same, and you would only notice a big difference if you ever tried to make your own package for your own software.

That said, a few package managers are very different from all the rest:

• Crux OS "prt-get": simple and stupid: just downloads and installs tar archives.
• Gentoo "emerge": builds all software from source code when you install it. This provides some guarantees that the source code was not tampered with by the distro maintainers, this is great if you need to review all of the source code that is running on your system, but terrible for most people who don't want to spend so much computing power on compiling stuff every time you do a software update.
• Nix and Guix: creates its own blockchain-like database of isolated package dependency chains on your system, allowing you to instantly roll-back to the previous set of installed packages if you ever install something that breaks your system. It also guarantees that the software can be checked bit-for-bit (using SHA hash) traced back to the exact version and dependencies of the source code that built it. Nix and Guix packages also live peacefully side-by-side with any other package manager since all Nix/Guix apps are completely self-contained within its own database. In a way, it is sort of like one big AppImage or Docker container, but you can just keep adding or removing stuff to it as often as you want.
• Silverblue, SteamOS, VanillaOS, BlendOS, CarbonOS: distributes "immutable images," so it is impossible modify the operating system at all. Updates will ship an entirely new operating system with all packages built-in. However you are allowed to install software into your home directory, and you can install FlatPacks and AppImages. This provides a great deal of security in exchange for a tiny bit of inconvenience.

My personal preference: I use ordinary Debian or Ubuntu to install the critical software that needs to be stable and reliable, and I use Guix OS on the side to install the bleeding-edge things that might break a lot.

The package manager is really only a small part of the story.

A distro at the end of the day is a API/ABI platform. What makes Debian what it is, is that it has a specific set of old unmoving packages. What makes Arch is that it has the latest APIs always. And everything in between like Fedora.

So even if Fedora used dpkg it wouldn’t change anything, you can’t use its packages on Debian.

As to why so many exist… well a lot of them suck in their own unique way.

I think a lot of what drives the creation of redundant open source tools is that the urge to address a matter of personal taste meets the urge to start a new project, so people create new things that are different in key ways from older ones, but not necessarily better, and not necessarily even different enough to justify the amount of work that goes into them.

In some ways it feels a lot easier to start a new project then to build off an existing one:

• You don't have to familiarize yourself with the old code, which may be in a language you don't know or don't like

• You don't have to deal with the existing maintainers, who may or may not be supportive of the changes you want to make

• You don't have to support use cases that don't matter to you personally

Some differences can be explained. Pacman was created after the Debian package manager (I guess that because Debian is older than Arch) . It is justified because Pacman is faster than Apt. But its too much work to replace Apt by Pacman comparing to the benefits.

But in some cases I don't know why. As instance I wonder why a distro, such as Void, created its own package manager instead of using the Alpine one. If Alpine is younger than Void, invert the sentence of course.

Because linux runs on an incredible amount of platforms with an incredible amount of hardware targets and platform goals.

It runs on every supercomputer and every raspberry pi and every android phone, as well as most web servers and almost every steamdeck.

Use cases are so dramatically different there will always be multiple distributions with their own needs for package managers.

Because people will never agree on a single one, and it's FOSS so nothing is forced. I for one am glad I don't have to use apt because I prefer pacman, just as I am glad someone who doesn't want to use an Arch-derivative has Debian and apt to fall back on.

Diversity generates competition that drives innovation. Also, people mostly work on what they like and use what they like. Although it certainly spreads resources on duplicate projects, you can't force people to work or use your personal favourite. You'll get used to it.

There are many different "X" because people have different tastes in their choice of "X". I like KDE, the next guy likes Gnome. I like Apt, but I might like whatever NIXOS uses, others like Yum or DNF. I kinda like the idea behind GoboLinux, probably because I was a MAC OSx user for a long while.

Any "why are there too many X's on Linux" (where X is package manager, desktop environment, init system etc) appear to stem from the silly assumption that there happens to be an already built operating system called Linux and all these people are forking it and putting in their own stuff for the sake of their own egos and nothing else.

When really, the answer fundamentally boils down to either one of two things: either it doesn't exist yet, or the existing solution fails to meet a need. Linux, itself, is merely a kernel; it didn't come with a package manager or desktop environment. Those things all had to be made by separate parties and there isn't always agreement on how best to do them.

As a Guix user, I believe the Guix package manager has advantages over "traditional" GNU/Linux package managers, as well as other so-called "universal package managers" such as Flatpak.

Using Arch in a distrobox, no difference, but the AUR is convienient.

Are they so different that it’s justified to have so many different distributions?

Linux isn't a project its a source compatible ecosystem. A parts bin out of which different people assemble different things. The parts being open source means you don't need anyone's permission or justification to make something different out of them.

From these many and varied efforts comes life, vitality, interest, intellectual investment. You can't just take the current things you like best and say well what if we all worked on THOSE when many of them wouldn't even have existed save for the existence of a vital ecosystem that supported experimentation and differentiation.

If we really believed in only pulling together maybe you would be developing in cobol on your dos workstation.

I don't think package managers are DE-only but rather depend on the distro

Some of them are more or less historic (for example rpm/dnf and dpkg/apt), where they were developed at a similar time by unrelated projects and have just carried on separately ever since.

Others have been developed to represent very different approaches (such as portage, which is based on the traditional BSD way of managing software by building from source, or snap and flatpak, which containerise applications).

The multitude of systems don't really cause as many problems as you'd think. As a rule, non-containerised packages need to be custom-built for each distro anyway, so it doesn't really make any difference which packaging tool is chosen by that distro. That is, you can't really take a debian .deb package and expect it to work properly on Fedora, even if you install dpkg/apt first.

People have different opinions on how packages should be managed. Of course, there are some package managers which are very similar to each other (DNF and zypper have the same backend), but they can also get really different (Nix/Guix and pacman are basically completely opposite in philosophy). It comes down to preference, and you can't force anything.

KDE and Gnome are just desktop environments. I use Gnome on Arch which uses pacman as the package manager. I could use KDE or XFCE or i3 or anything. The package manager is tied to the distro, and you can use almost any software with them.

“Are they….justified”?

1. Somebody thought the need for a new package manager was great enough to spend time creating one. That person at least must think it is justified.

2. We, the users, have not chosen just one of the options to be the standard. Does that “justify” that they all exist?

In the short term, the popularity of Linux is certainly hurt by the complexity of the ecosystem and the lack of standardization. As a product, it would see better adoption of it were more standardized. Without writing a book about why, there is no doubt about this. The short version is that, today, Linux is many products, none of which can compete as effectively as one would and all of them are impaired by the confusion this causes.

In the longer run though, it is almost certainly one of the great strengths of Linux. Linux is many products and as a result, it can target and effectively fill almost every niche. That is going to make it very hard for alternatives to compete at some point. Once Linux knowledge and Linux applications ( yes, I know ) become more mainstream, this compatibility between options becomes a strength. I can have my own operating system that is just the way I want it, but it still runs Docker and Stream ( as examples ).

Think of the cereal aisle at the grocery store. If I want to introduce a new cereal ( or pasta sauce or whatever ), coming up with one that has 10 flavours is not going to work ( without immense marketing muscle ). None of them will sell well enough and probably all of them will get pulled from store shelves. I would be better off launching one. However, once I have a mature market position, I can have not just the regular version but the whole wheat version, the honey nut version, the cinnamon version, the holiday version , etc. They will collectively make each other stronger and all potentially sell well ( again, think pasta sauce flavours if that makes more sense to you ).

This is why there was The Tesla Roadster at first and now there are the Model S, Model 3, Model X, Model Y, and maybe the Cyber Truck.

Linux is not a “product” though. It is an Open Source program. While any given Linux distributor ( distribution ) may think like I outline above, collectively the Linux market is fragmented. Linux is a mix of commercial, community, and individual interests all scratching their own itch.

I am super interested in Chimera Linux right now and fairly negative towards Ubuntu. This makes me part of your problem though. Chimera Linux makes “Linux” less predictable, more confusing, and more frustrating for new and potential users. Pushing everybody to Ubuntu would be a better market strategy. That said, I personally want to use Chimera Linux and, while I say that I want Linux to succeed, I also secretly hope that Ubuntu will fail. Chimera Linux uses a package manager used by only one other Linux ( and in fact they use different, incompatible versions of it so really they are unique ). Clearly, my priorities are mid-aligned with the premise of your question.

So, what does “justified” mean in the Linux space.

Imagine having only one option, and that option is dnf. I'm out. I don't want Linux anymore.