Post B4yha7xWnaT6EApnKS by domi@donotsta.re
 (DIR) More posts by domi@donotsta.re
 (DIR) Post #B4yZ4kRlMs4eh9VEps by asie@mk.asie.pl
       0 likes, 1 repeats
       
       https://s3kshun8.games/blog/flatpak-won/This should be mandatory reading for everyone invested in the Linux ecosystem, whether you leave feeling heard by or enraged by the author.
       
 (DIR) Post #B4ybRajFrfVH7lhl3I by asie@mk.asie.pl
       2 likes, 0 repeats
       
       @toast@donotsta.re I agree that the post does presume an investment in the "growth of the Linux ecosystem". If one sees distribution/operating system/packaging work as a creative endeavor for its own sake rather than an utilitarian craft for usage by and/or empowerment of the masses, then most of the criticism will naturally bounce off as unrelated.
       
 (DIR) Post #B4ydSPwzcjs3tSeVPM by wolf480pl@mstdn.io
       0 likes, 0 repeats
       
       @asiesorry I quit after the first sentence
       
 (DIR) Post #B4yha7xWnaT6EApnKS by domi@donotsta.re
       0 likes, 0 repeats
       
       @asie As a penguin, I was exposed early on to the ideas of Stallman. I respect and agree with what he believes in.fossbro, opinion discarded /si think the author fails to actually touch on the core of any of the issues he’s writing about :/while Flatpak users across the ecosystem and on all distributions casually download things off the Discover store, expect them to Just Work, and actually that’s exactly what happens every single time.HALF OF THE TIME I DOWNLOAD SHIT FROM FLATPAK IT DOESN’T FUCKING WORK. not to even mention how much it sucks ASS for console applications. like, sure, i don’t want to erase the author’s experience, i bet it works okay for his userbase, but he’s generalizing.also the casual hate on nix, explaining issues you’d never face unless you explicitly decided to start using nixos. what the fuck man
       
 (DIR) Post #B4yiKiQnWpsefGjwuG by wolf480pl@mstdn.io
       1 likes, 0 repeats
       
       @asie to elaborate:> As a software engineervery few programmers would qualify as actual engineers, certainly not me> the most important thing in the universe is your ability to actually get people to experience your work.No.If we take "engineer" literally, the most important thing is to prevent your work from putting users in danger. Sometimes that means not releasing your software.If we ignore the "engineer" part, I still think there's too much software already.
       
 (DIR) Post #B4yjFspTXdP3Q5XhOy by asie@mk.asie.pl
       0 likes, 0 repeats
       
       @domi@donotsta.rei think his position in that regard is coming from that of a software maintainer. if my opinion on NixOS was based solely on the disappointment in users' faces when I told them "haha, good luck" when asked about how to make Wonderful Toolchain run on it, i would have a pretty low opinion of it as wellI remember when some software projects banned bug reports from Gentoo because the combination of USEFLAGS, unusual configurations, and tinkering-happy users made it impossible to reproduce anything
       
 (DIR) Post #B4yjHFzgtv51YNyYcq by asie@mk.asie.pl
       0 likes, 0 repeats
       
       @domi@donotsta.re (yes yes I know NixOS solves exactly the problem of it being impossible to reproduce anything)
       
 (DIR) Post #B4yjbEzAMNquDYG8K8 by aura@gts.foxsnuggl.es
       0 likes, 0 repeats
       
       @asie the most contentious bit in that post is that FOSS maintainers would've prevented the crowdstrike incident when the only mechanism by which that is true is the shitty state of packaging that is also being criticizedsee xz backdoor being noticed in debian unstable
       
 (DIR) Post #B4yjbFK56caBGQ2pvM by asie@mk.asie.pl
       0 likes, 0 repeats
       
       @aura@gts.foxsnuggl.es you know that's a pretty good point actually
       
 (DIR) Post #B4yk6ZtKgMgO27TPUG by asie@mk.asie.pl
       1 likes, 0 repeats
       
       @domi@donotsta.re (also someone did make Wonderful Toolchain run on Nix!https://github.com/pgattic/blocksds-nix/blob/master/flake.nixwell BlocksDS but I wouldn't mind someone adapting it to do the WonderSwan part)
       
 (DIR) Post #B4ymuHDCbeKOUGaq4O by tea@snug.moe
       0 likes, 0 repeats
       
       @asie I think this post is good in the wake of lots of maintainers of (gui) applications being against flatpak. I want to run a simple command to install your application, and not download an appimage which likely never ends up leaving my downloads folder. I want automatic updates for those applications (whenever I decide to run flatpak update) instead of them doing their own update checks and bothering me with popups on startup. I get that publishing to flathub is extra effort, but why is barely anyone building .flatpaks for their releases, when it is "close" downloading and running an appimage?
       
 (DIR) Post #B4ymuHX3PqCvTpsh0q by asie@mk.asie.pl
       1 likes, 0 repeats
       
       @tea@snug.moe There are some programs which can't be published to Flathub, unfortunately. In particular, I'm thinking of Minecraft launchers, where old versions of the game don't run well on Wayland at all, while new versions of the game run worse on xwayland than native Wayland:https://github.com/flathub-infra/flatpak-builder-lint/pull/935
       
 (DIR) Post #B4ymzNxuvNSOsesbku by asie@mk.asie.pl
       0 likes, 0 repeats
       
       @tea@snug.moe (Well, Pandora is on Flathub, but...https://github.com/Moulberry/PandoraLauncher/commit/1db9c06e145796a5eed1020638010768e78198c1#diff-c8f792921248386552ff596b903f7cd4bb20339255d0a8d674619d6d72bda984R209)
       
 (DIR) Post #B4yvmhTRyoOfCyegGu by autumn@translunar.academy
       0 likes, 0 repeats
       
       @asie I read it and came away thankful that I stopped writing code for others several years ago. 🤷‍♀️I still write code though, and put it out there, but had to walk away from the “community” part.
       
 (DIR) Post #B4z09j2AkHDgWX8AjY by xgqt@functional.cafe
       0 likes, 0 repeats
       
       @asie If we were to agree with author I find it hard to "exercise the four freedoms" (https://www.gnu.org/philosophy/free-sw.en.html#four-freedoms) with some of their points.
       
 (DIR) Post #B4z09jFHxViBBDGeB6 by asie@mk.asie.pl
       0 likes, 0 repeats
       
       @xgqt@functional.cafe I think the author's trying to point out that it is hard for most users of GNU/Linux to "exercise the four freedoms", or in fact benefit from other people exercising those freedoms in a practical sense today; in this way, the offering of the corporate machine becomes more compelling.Having the freedom to do something does not absolve its user of criticism.
       
 (DIR) Post #B4zLcVYa1xx89yr5EG by a@layer02.net
       0 likes, 0 repeats
       
       @toast @asie i don't mind if devs distribute their own apps, but that absolutely shouldn't be the *only* way to get the apps. it's a simple proposition: i don't trust all devs, but i do trust my distro maintainers. apps that assume/require a certain distribution method are apps that i simply will never use. moreover, distribution methods that assume/require everything else to use the same distribution method are going to be similarly problematic. (practical example: obs-studio built by arch doesn't have the twitch/youtube/etc api integration, but obs-studio flatpak doesn't work with blackmagic decklink drivers, but obs-studio-tytan652 in the aur just works.)the far bigger problem i take with the article, other than the tone being really unpleasant, is that one of its central claims is that increased centralization is a good thing. first of all, it's a complete non-sequitur to why appimage sucks (and it does suck). beyond that, i don't think the historical framing is accurate either -- hostile software is not a thing of the past, "well-paid engineers that act in good faith" aren't the only participants nor are they a guarantee that problems won't occur, and "compilation and distribution [being] more centralized [...] than [they've] ever been" is not an unquestionably good thing. "Towards improvement, towards centralization" sounds like an oxymoron unless you accept that centralization is aligned with improvement, which... there are many reasons *not* to accept that???? when the strawman being constructed is "you're somehow above just using the developer's build of the developer's application they provided to you", as if there aren't *any* valid reasons to take any sort of issue with those builds?but also the tone, my god:> The guys out there with big Che Guevara energy are the real ones building and perpetuating a misery machine fueled by your ideology and nothing else. It's a wealth transfer from the hearts and minds of real and talented developers to ideologues at the top of the ecosystem that would rather take value from your work whilst not actually contributing anything and hoping you'll join in singing kumbaya, drinking the kool-aid, and starving to death while you do one of the hardest jobs on earth.like, hoo boy, for someone complaining about "ideology" this paragraph alone is full of ideology> This is incontrovertible fact you can only fail to see if, to you, Linux is not the democratization of technology but instead the intimate satisfaction of init scripts sliding up your butthole in just the right way.which i guess illustrates the fundamental flaw of this whole article, that it complains about one thing but uses it to somehow justify something completely different and unjustifiable. what is "the democratization of technology" to this person? is it "democratization" when all the technology is controlled entirely by whatever gets corporate funding, and anything that isn't valuable to those corporations remains unpaid? should linux be this club run by those "well-paid engineers" they seem so fond of? how is "democratization" compatible with "centralization"?in conclusion i didn't get anything useful out of reading this article and whatever useful points could have been salvaged from this, i'm sure others have made those points in far more coherent arguments :cirno_shrug: