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.
-
@glyph ... Why is the redundancy a problem?
I've generally considered it a feature of the linux ecosystem that Serious Projects
like debian share an ecological niche with shit like that Hannah Montana distro that briefly went viral probably a decade ago.@miss_rodent Multiple distros means multiple platforms. Multiple platforms means multiple dev targets. Multiple dev targets means that developing for "Linux" is, at minimum, *many dozens of times more expensive* than developing for macOS, or Windows, or the web.
-
@miss_rodent Multiple distros means multiple platforms. Multiple platforms means multiple dev targets. Multiple dev targets means that developing for "Linux" is, at minimum, *many dozens of times more expensive* than developing for macOS, or Windows, or the web.
@miss_rodent The important thing is to get the "paying end-user / cost to target platform" ratio down below a threshold that resembles macOS. An actually profitable platform like iOS or Windows is unrealistic, but macOS is a minority with enough users for devs to care about without bankrupting themselves and at least build a lifestyle brand around.
-
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 Completely disagree. The fact that we don't have a "real Linux platform" is our strength. It's non-monoculture. It's why software has to be written to follow specifications rather than treating an implementation as the specification. It's why the BSDs etc. are still viable too - software that's portable to different Linuxes is usually also portable to them. It's why we're not stuck listening to the worst people forcing their ideas on us, but can make something different that still runs basically all the same software when you want it to.
-
@miss_rodent Multiple distros means multiple platforms. Multiple platforms means multiple dev targets. Multiple dev targets means that developing for "Linux" is, at minimum, *many dozens of times more expensive* than developing for macOS, or Windows, or the web.
@glyph @miss_rodent But things like Flatpak exist.
-
@miss_rodent Multiple distros means multiple platforms. Multiple platforms means multiple dev targets. Multiple dev targets means that developing for "Linux" is, at minimum, *many dozens of times more expensive* than developing for macOS, or Windows, or the web.
@glyph @miss_rodent Multiple platforms are only multiple dev targets if you're in a 1980s-mindset of #ifdef hell everywhere. If instead you write portable code to the specs not to "works on my machine", and probe for things rather than enumerating a million build combinatorics, supporting a multitude of different systems is not any harder than supporting one.
-
This is a big problem because labor is not fungible and the reason that a lot of these people got involved in distro development *in the first place* is that those sorts of problems and systems are interesting and engaging for them to work on. They all want to have control over a packaging tool, or a build farm, or whatever. The redundancy is a huge problem and a huge waste but it's also the engine that powers the whole thing, to some extent.
@glyph mmm, not just because of "interesting and engaging problems", but many people came into it specifically because of their egos: *my* distro will be better (and I will be famous). Basically, you're advocating for abandoning the core motivation (or one of the main ones anyway) for tinkering around the free code.
-
@miss_rodent The important thing is to get the "paying end-user / cost to target platform" ratio down below a threshold that resembles macOS. An actually profitable platform like iOS or Windows is unrealistic, but macOS is a minority with enough users for devs to care about without bankrupting themselves and at least build a lifestyle brand around.
@glyph Honestly, that also seems fine to my perspective?
People still develop for 'linux' in various capacities. It makes it dificult to compete on a commercial/industry level...
but also, that is not why I'm in this ecosystem in the first place - I'm mainly in the linux ecosystem because fuck the capitalistic software model completely?Finding a way or people to better survive & eat within this ecosystem I think is a worthwhile goal - but not one worth the cost of consolidation
-
@glyph @miss_rodent Multiple platforms are only multiple dev targets if you're in a 1980s-mindset of #ifdef hell everywhere. If instead you write portable code to the specs not to "works on my machine", and probe for things rather than enumerating a million build combinatorics, supporting a multitude of different systems is not any harder than supporting one.
@dalias @miss_rodent spoken like a Java marketing executive from 2002. it will definitely be cool if this ever happens but after 30 years of waiting for it to materialize, I think we can call the experiment a failure
-
@glyph Completely disagree. The fact that we don't have a "real Linux platform" is our strength. It's non-monoculture. It's why software has to be written to follow specifications rather than treating an implementation as the specification. It's why the BSDs etc. are still viable too - software that's portable to different Linuxes is usually also portable to them. It's why we're not stuck listening to the worst people forcing their ideas on us, but can make something different that still runs basically all the same software when you want it to.
RE: https://hachyderm.io/@dalias/116189951658156876
@glyph And I know you're well-meaning, but this kind of wish to tear up the non-monoculture we've so painstakingly nourished over the decades, fighting to preserve consensus standards process and portability, is a big part of:
-
@dalias @miss_rodent spoken like a Java marketing executive from 2002. it will definitely be cool if this ever happens but after 30 years of waiting for it to materialize, I think we can call the experiment a failure
@glyph @miss_rodent It does work. This is why basically everything that was originally written for GNU/Linux compiles out of the box without problem on Alpine with musl. It was sometimes hard work fixing old bad practices people were stuck on, but now everyone benefits. And it HURTS when somebody wants to tear that up.
-
@glyph Honestly, that also seems fine to my perspective?
People still develop for 'linux' in various capacities. It makes it dificult to compete on a commercial/industry level...
but also, that is not why I'm in this ecosystem in the first place - I'm mainly in the linux ecosystem because fuck the capitalistic software model completely?Finding a way or people to better survive & eat within this ecosystem I think is a worthwhile goal - but not one worth the cost of consolidation
@miss_rodent the consolidation is going to happen anyway. I am suggesting a way it could happen with democratic involvement of volunteers. which I realize is a bit of a pipe dream.
realistically, everyone will just pivot what it means to make "linux desktop" software to mean "works on SteamOS" and then Valve gets to write the specs that everyone else follows, and the viability of a desktop Linux distro will be scored according to the accuracy of its SteamOS emulation
-
@glyph @miss_rodent But things like Flatpak exist.
@cthos @miss_rodent do they? experts seem to disagree
-
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 And there are already plenty of people who feel that systemd was imposed against their will. No, I feel like this can't be the right answer. Especially when the problem, as I understand it, is that the Linux bazaar is uncomfortable to those of us who primarily ship applications for other, proprietary platforms. Why would they want to fundamentally change what they're doing to fit our conception, which they presumably view as flawed, of what a platform should be?
-
@cthos @miss_rodent do they? experts seem to disagree
@glyph @miss_rodent I mean, I run several different distros across several different pieces of hardware, with different desktop environments and I can install Flatpaks on all of them, so yes?
-
@glyph And there are already plenty of people who feel that systemd was imposed against their will. No, I feel like this can't be the right answer. Especially when the problem, as I understand it, is that the Linux bazaar is uncomfortable to those of us who primarily ship applications for other, proprietary platforms. Why would they want to fundamentally change what they're doing to fit our conception, which they presumably view as flawed, of what a platform should be?
@matt they're free to read my analysis, think "nah", and forget about it. I'm not particularly influential here, it just bothers me.
but the reason it bothers me is that the cost here is permanent irrelevance. is the point of a free software desktop to have direct, monarchical control of the development process of your compositor or whatever, or is it to provide *users* with a more accessible and open computing experience where they can have agency and control over their applications?
-
Fixing the problem involves driving a truck through that load-bearing "to some extent". Doing a big ugly multi-party negotiation to figure out how we can EOL Qt, to replace it with Gtk everywhere, and get all the Gtk devs on board with being *extremely* nice to the Qt people as we sunset their work. (Did you feel a little thrill because I picked Gtk instead of Qt? Well, I flipped a coin. Imagine I said Qt wins instead of Gtk. You're gonna be that mad about *big* parts of this, no matter what.)
@glyph so, to invent linux need to invent society first?
-
@matt they're free to read my analysis, think "nah", and forget about it. I'm not particularly influential here, it just bothers me.
but the reason it bothers me is that the cost here is permanent irrelevance. is the point of a free software desktop to have direct, monarchical control of the development process of your compositor or whatever, or is it to provide *users* with a more accessible and open computing experience where they can have agency and control over their applications?
@matt I would like to see the efforts of the movement directed to the kind of liberatory project that everyone involved in "Free Software" pretends to care about, for actual users, not just themselves. which means making it popular enough that it is the system that kids accidentally encounter at school and at home and at the library. which means making it the system that people get on their work computers.
-
@glyph @miss_rodent I mean, I run several different distros across several different pieces of hardware, with different desktop environments and I can install Flatpaks on all of them, so yes?
@glyph @miss_rodent The list includes but is not limited to:
- Manjaro on a 2015 Macbook Air 11" with XFCE
- Bazzite on a Framework 13" with KDE
- ZorinOS on Starlabs Starlite (which IIRC is highly skinned GNOME)
- Vanilla Ubuntu on a weirdo 10" tablet PC thingie from Chuwi (Required some config to enable because Ubuntu really loves snaps and they shouldn't)And all my applications just work.
-
@glyph @miss_rodent I mean, I run several different distros across several different pieces of hardware, with different desktop environments and I can install Flatpaks on all of them, so yes?
@cthos @miss_rodent less flippant answer:
there are of course efforts to unify the platforms around certain abstractions which paper over the differences. and some of them (flatpak included) are even close enough to kinda work some of the time. but developing a flatpak and getting it deployed, while *possible*, does not have zero marginal cost per distro.
-
@cthos @miss_rodent less flippant answer:
there are of course efforts to unify the platforms around certain abstractions which paper over the differences. and some of them (flatpak included) are even close enough to kinda work some of the time. but developing a flatpak and getting it deployed, while *possible*, does not have zero marginal cost per distro.
@cthos @miss_rodent and to the extent that it succeeds, it succeeds by creating a meta-platform, flatpak, that floats on top of the distro and makes all the distinctions between them irrelevant anyway. it also doesn't fully succeed (flatpak filesystem permissions are a user-interface nightmare)
