I'm mad about linux distros again today and I think I am realizing why this is so hard for me to write about systemically: I have a software engineer brain and so I try to model the various problems as technical problems.
-
@cthos @miss_rodent I think I do have some nuanced structural critique of flatpak that I sadly don’t have time to get into right now, but if I am being honest most of my systems have a weird quirk where user data lives outside home directories on external media and this causes flatpak’s weird slightly-wrong but makes-things-mostly-work heuristics absolutely violently explode in ways which cause huge issues that contributes to an overall *immediate* negative impression
@glyph @miss_rodent Understandable, and yes, there are some pretty sharp edges.
-
In short, all the volunteer-based distributions need to have a gigantic conference where they all come together and *agree to stop working on about 99% of them*, to pool efforts to make a real Linux platform. A lot of people will need to put their egos aside and decide to acquiesce to solutions they believe to be technically inferior, in order to be able to address the diffusion of labor into pointlessly recreating basically the same toolchain a thousand times.
@glyph but why? What's the problem if there's a long tail of distributions with few developers and users, in addition to the big ones? Some may have only minor differences, but many focus on a specific niche or are built around completely different technologies that solve different problems, and you never know which ones will become important in the future, either by themselves or by proving that something works well enough to be widely adopted
-
@glyph @miss_rodent But things like Flatpak exist.
@cthos @glyph @miss_rodent I can't speak to macOS since I don't own an Apple device and thus don't have access to any of that world (maybe I'll pick up a Neo if I have a few spare bucks so that I can develop for Arcalibre there), but Windows is far from a monolithic platform these days.
It's also notable that I can use Windows to develop for Linux, given Docker and WSL, but it's much harder to use Linux to develop for Windows due to entirely manufactured obstacles.
-
@cthos @glyph @miss_rodent I can't speak to macOS since I don't own an Apple device and thus don't have access to any of that world (maybe I'll pick up a Neo if I have a few spare bucks so that I can develop for Arcalibre there), but Windows is far from a monolithic platform these days.
It's also notable that I can use Windows to develop for Linux, given Docker and WSL, but it's much harder to use Linux to develop for Windows due to entirely manufactured obstacles.
@cthos @glyph @miss_rodent That is, I incur a significant additional expense developing for Windows as compared to developing for Linux — an expense folks have been kind enough to help with, but an expense nonetheless.
The expense isn't just a proliferation of distros, it's also how easy it is to access and use tools for dealing with that proliferation.
-
In short, all the volunteer-based distributions need to have a gigantic conference where they all come together and *agree to stop working on about 99% of them*, to pool efforts to make a real Linux platform. A lot of people will need to put their egos aside and decide to acquiesce to solutions they believe to be technically inferior, in order to be able to address the diffusion of labor into pointlessly recreating basically the same toolchain a thousand times.
@glyph (fully recognizing how helpful the following take isnt)
Yes, but also extremely no.Being able to put forward a coherent, open-source, not-megacorp-owned product that is approachable for everyday use probably only happens if we do something like this.
On the other hand, there is strength and value in decentralization, and also value in specialist and niche distributions (even if some of their value is simply the delight they gave their developers)
I would also bet that a at least half of the stubborn split efforts are not around technical merits or the way they tie into developers egos.
-
@cthos @miss_rodent FWIW it's not *impossible* for this to work, but it is wildly beyond *cost-effective* for most ISVs
@glyph @cthos @miss_rodent Then the community should definitely organize to make Flatpaks work more reliably across distros. What are the biggest problems that make it necessary for app developers to put in distro-specific work even when targeting Flatpak? (I have no experience with this yet; my main desktop app project at this point is a remote desktop access tool, so that's a whole other can of worms.)
-
@cthos @glyph @miss_rodent That is, I incur a significant additional expense developing for Windows as compared to developing for Linux — an expense folks have been kind enough to help with, but an expense nonetheless.
The expense isn't just a proliferation of distros, it's also how easy it is to access and use tools for dealing with that proliferation.
@cthos @glyph @miss_rodent Linux isn't a platform because it's too many different platforms under one name, but Windows isn't a platform because it fails to be a platform at all. Many and zero both fail to be one.
-
@glyph (fully recognizing how helpful the following take isnt)
Yes, but also extremely no.Being able to put forward a coherent, open-source, not-megacorp-owned product that is approachable for everyday use probably only happens if we do something like this.
On the other hand, there is strength and value in decentralization, and also value in specialist and niche distributions (even if some of their value is simply the delight they gave their developers)
I would also bet that a at least half of the stubborn split efforts are not around technical merits or the way they tie into developers egos.
@glyph (rereading, I realize you did not say or even imply that we needed to unite around *one* distro/project, but rather we need around 99% fewer. In that case, maybe I'm just quibbling over whether 99% is the right number or if maybe it's more like 80%)
-
@cthos @miss_rodent I think I do have some nuanced structural critique of flatpak that I sadly don’t have time to get into right now, but if I am being honest most of my systems have a weird quirk where user data lives outside home directories on external media and this causes flatpak’s weird slightly-wrong but makes-things-mostly-work heuristics absolutely violently explode in ways which cause huge issues that contributes to an overall *immediate* negative impression
@glyph @cthos @miss_rodent Sorry, hadn't yet seen this post.
-
In short, all the volunteer-based distributions need to have a gigantic conference where they all come together and *agree to stop working on about 99% of them*, to pool efforts to make a real Linux platform. A lot of people will need to put their egos aside and decide to acquiesce to solutions they believe to be technically inferior, in order to be able to address the diffusion of labor into pointlessly recreating basically the same toolchain a thousand times.
@glyph this assumes I have interest in (volunteer) working on something watered down enough to work for 99% of people
-
@glyph this assumes I have interest in (volunteer) working on something watered down enough to work for 99% of people
@glyph I like building a small garden not industrial agribusiness
-
In short, all the volunteer-based distributions need to have a gigantic conference where they all come together and *agree to stop working on about 99% of them*, to pool efforts to make a real Linux platform. A lot of people will need to put their egos aside and decide to acquiesce to solutions they believe to be technically inferior, in order to be able to address the diffusion of labor into pointlessly recreating basically the same toolchain a thousand times.
@glyph It’s not just about egos either. Different people have different incompatible ideologies. You have free (libre) software purists and open source purists and free (beer) purists and license hawks and IP anarchists and people who may not be able to say it out loud but are quite happy to serve the interests of capital and people who can’t even agree what serves the interests of capital. And many of these aren’t differences of opinion in means but differences of opinion in ends. You can’t get people to agree who don’t agree on goals. If you can get everybody to agree that the goal is to create the Linux, you have a shared goal and it is conceivably achievable. I can’t see that happening. Distributions, like all software, are always a means to an end.
-
@glyph It’s not just about egos either. Different people have different incompatible ideologies. You have free (libre) software purists and open source purists and free (beer) purists and license hawks and IP anarchists and people who may not be able to say it out loud but are quite happy to serve the interests of capital and people who can’t even agree what serves the interests of capital. And many of these aren’t differences of opinion in means but differences of opinion in ends. You can’t get people to agree who don’t agree on goals. If you can get everybody to agree that the goal is to create the Linux, you have a shared goal and it is conceivably achievable. I can’t see that happening. Distributions, like all software, are always a means to an end.
@glyph That might be the downfall of Linux, but isn’t there a certain beauty in that too? You don’t have to use a system created by people who you hate. Or worse and more likely, who hate you.
-
@glyph @cthos @miss_rodent Then the community should definitely organize to make Flatpaks work more reliably across distros. What are the biggest problems that make it necessary for app developers to put in distro-specific work even when targeting Flatpak? (I have no experience with this yet; my main desktop app project at this point is a remote desktop access tool, so that's a whole other can of worms.)
@matt @cthos @miss_rodent flatpak does not address the deficiencies or the expense of interacting with a zillion different compositors, or for that matter different audio systems or GPS daemons. the kinds of apps that need to interact with the platform need an API shaped in terms of d-bus endpoints, and the problem that flatpak addresses is one of .so files. flatpak also requires manual management of filesystem permissions, which means apps are just slightly dysfunctional
-
@matt @cthos @miss_rodent flatpak does not address the deficiencies or the expense of interacting with a zillion different compositors, or for that matter different audio systems or GPS daemons. the kinds of apps that need to interact with the platform need an API shaped in terms of d-bus endpoints, and the problem that flatpak addresses is one of .so files. flatpak also requires manual management of filesystem permissions, which means apps are just slightly dysfunctional
@matt @cthos @miss_rodent the main thing is that it is missing a macOS style powerbox file dialog (which, ironically, was originally invented in linux, via bifrost in OLPC)
-
@glyph I like building a small garden not industrial agribusiness
@jay this is what homelabbing is for, not operating system development
-
@matt @cthos @miss_rodent the main thing is that it is missing a macOS style powerbox file dialog (which, ironically, was originally invented in linux, via bifrost in OLPC)
@matt @cthos @miss_rodent I think *most* telling though, the thing that is downstream from all the various subtle upstream problems with flatpak, is why does the platform still have “native” apps and “flatpak” apps as separate categories? is there an OS yet which is ONLY a runtime for flatpaks and doesn’t have a privileged class of “good” apps which don’t have to live in flatpak jail?
-
@matt @cthos @miss_rodent the main thing is that it is missing a macOS style powerbox file dialog (which, ironically, was originally invented in linux, via bifrost in OLPC)
@glyph @cthos @miss_rodent It looks like the File Chooser portal (https://flatpak.github.io/xdg-desktop-portal/docs/doc-org.freedesktop.portal.FileChooser.html) is supposed to implement a powerbox file dialog. I assume it somehow doesn't get this right.
-
@matt @cthos @miss_rodent I think *most* telling though, the thing that is downstream from all the various subtle upstream problems with flatpak, is why does the platform still have “native” apps and “flatpak” apps as separate categories? is there an OS yet which is ONLY a runtime for flatpaks and doesn’t have a privileged class of “good” apps which don’t have to live in flatpak jail?
@glyph @matt @miss_rodent I don't agree that this is a "problem" per se, but most of the immutable distros (SteamOS, Bazzite, VanillaOS) lean _heavily_ on Flatpaks because you cannot just do `apt install {foo}`.
-
@matt @cthos @miss_rodent I think *most* telling though, the thing that is downstream from all the various subtle upstream problems with flatpak, is why does the platform still have “native” apps and “flatpak” apps as separate categories? is there an OS yet which is ONLY a runtime for flatpaks and doesn’t have a privileged class of “good” apps which don’t have to live in flatpak jail?
@matt @cthos @miss_rodent as long as the native version of apps still exists then that’s still 1 platform with 2 platform targets which ISVs now have a bunch of research to do to find out users are gonna be mad if they package their particular app like that. which by itself is already more work. and if your platform also supports whatever a “snap” is, god help you
