No matter how fanboi-y a Linux or Apple user gets, they can never out fanboi a Microsoft fanboi. They take making shit up about competitors to a entirely new level.
Thats… Thats like a flat earther in computer stuff.
Do those really exist??
I know it gets thrown around a lot, but the Dunning-Kruger effect is real and applicable to people in all fields.

what a great way to create a set of echochambers with schizo worldviews
But once you read his words he’s got a foot in the door. Then he’s harder to ignore.
So maybe it’s harder to ignore fools on social media. Which would make social media a kind of fool-enhancer.
I guess this is where blocking comes in. But that seems drastic.
Blocking is severe, but boy is my feed clean of morons (I think I’ve only blocked like 30 people on Lemmy).
You gotta try it. Very satisfying to click ‘read all’ on your inbox now and then to clear out notifications for new (hidden) messages from trolls you’ve blocked.
sometimes if you have nothing nice to say to that person, just post rocky horror .gifs. i really wish this site would have a .gif finder though. seriously!
Lame attempt at ragebait.
Meanwhile Microsoft makes the start menu with React
I try not to let considerations get in the way of doing great work.

That is the most punchable response I’ve seen in a while.
Punchable is a bit far, probably wanna tone it down a bit, big guy.
Just kidding, but it’s only funny and also is it this guys fault?
I don’t even know if it’s true, but in any case, the guy who tasked a react (native) developer on the start menu is responsible (not the developer).
Example: If I managed a product and hired a python developer and told them to do x, they would likely use python, right? (In this scenario, It is I the manager wjo failed everyone, not the developer).
Also the other commenter is correct. It’s like the common saying “use the right tool for the job”. The saying doesn’t make sense, because the right tool is always the one you know how to use…
Punchable is a bit far
Didn’t mean that literally 😁 I just thought it was funny to describe a written response as punchable. But the response was annoying.
It might not be the developer’s fault, but him practically defending the choice by completely dodging the performance aspect is irresponsible to me. It’s like he’s ashamed to discuss it. Or doesn’t have the knowledge to?
A good developer should speak up and say that they might not have the skill set required for the task. That’s my opinion. Either that or learn the ropes but flag for extra time needed.
the right tool is always the one you know how to use…
100% disagree. There are more suitable tools for certain jobs.
Hello I am looking for a job as a surgeon. Okay what tools do you know? Jackhammer. That’s the only one you know? Yup. Mmkay then that’s the right tool for surgery in your case, can you start Monday?
🫠
Imagine someone using Java to write a program that just runs several other commands in sequence. That should really be a script, and the developer should learn a sufficiently suitable shell scripting language.
But yeah, ultimately I agree, it’s not his “fault”. Especially if he flagged this to be something a bit out of the ordinary, and his manager(?) insisted. Then it’s 100% not the guy’s fault, for sure ong.
When you have a hammer, everything looks like a nail.
I feel like that reply would have fit better one comment level higher.
Very possibly. Oh well…
“… And that is how I used React as a database language.”
~me presenting at a conference in 2026
“I am a react developer”
I thought we are supposed to be language-agnostic after 3rd project.
In the era of tech evangelists? People pick 1 technology branch and make it their entire personality
That’s still a stupid reason. I’m a .NET & MSSQL developer primarily, I’m not gonna shove C# in every project I write if it doesn’t makes sense.
Two more projects to go, then!
No way that’s real. Tell me it’s not real
Probably, but only because at this point I’m fairly certain reality itself must be a parody of something.
This is what’s wrong. Distilled.
Well, at least it’s React Native, seemingly. Also from what I’ve heard it’s only one section like rendering results from the web or some shit like that.
What kind of shit for brains asshole is still defending Windows in 2025?
And what kind of slavering mouth-breathing teoglodyte doesn’t understand that Hannah Montana Linux negates all of these issues, will suck your dixk without hesitation, and lets you read news from four days from now.
Lololol
Aren’t Qt and GTK cross platform? I have Dolphin and Kate running on my Windows work laptop.
Yeah, he doesn’t know what he’s talking about. There is a shitload of frontend developers that specialize in web standards and technologies. Electron was developed to take advantage of that deep pool of frontend developers. The side affect, is that other OSes can just support electron and they get the developers and the applications for free. Which has been a major boon for Linux users and those looking to escape Microsoft’s vendor lockin strategy. Today might be different, but in the past, nobody was intending to support Linux by creating electron apps. If they cared so much or it was so important, they would have been using Qt and GTK prior to Electron.
As is wxWidgets and others.
Dear god, I had to write wxWidgets/C++ for Win32 last year for work and it was horrific. Never again, back to modern web standards for me. The irony is that it was justified as being “cross platform” but we never got around to actually making it work on Linux. Makes no sense, it should have been an internal web app. (Admittedly, this was for law enforcement software and they seem to love windows.)
Tell me you have no idea how software development works without saying it…
Tbh, it’s not entirely wrong, which is the reason why it works so well as rage bait.
It’s really not about Linux, but it is about supporting anything and everything out there with a single app. Use Electron and you can have the same app running on Windows, Mac, Linux, Android, iOS, your car, your game console, your smart fridge and in a website.
Of course the result sucks, but if you can cut development effort into a fraction while also supporting systems that you would have never supported otherwise, that’s not a bad deal for businesses.
I read it twice and it felt like word salad.
Okay, how does this dude explain native Linux apps?
Probably “Native Linux apps are made in Linux-only bullshit by useless neckbeards, and probably only run in the terminal. Real actual apps like Discord made by a for-profit corporation have to be made cross-platform.”
Electron is the only cross platform gui toolkit…
If you ignore QT, GTK and everything else.
I’m so glad that Microsoft makes an awesome cross platfor— wait, no, but they contribute code to— hmmm … Hey, what does Microsoft do to make apps more portable again?
The real reasons often are:
- They want be able to hire much cheaper webdevs instead of software devs.
- Electron has a lot of built-in data collecting metrics, which they urgently need for creating a real-life KITT.
- Easy live embedding of content. Sure you can add your own solution, in fact I created ETML as a solution for this problem for my engine, all without any support for nasty scripting languages or convoluted stylesheets (style-inheritance in CSS turned me off from webdev even more than JS did). At best, it can be used for things like embedding videos on Discord, because no one else thought some universal approach, let alone one that disallows proprietary players. At worst, it’s being used for ads.
Also a lot of Windows-only apps are Electron apps, only because the manufacturer wants to go “fuck you”, even putting protections into the code just in case you wanted to run it on Linux.
EDIT: Forgot the “live embeds” reason.
Another reason is when developing the Web version first. Draw.io is a good example, where we get a bonus desktop(electron) version “for free” though the product was developed as a web app.
Also works on mobile easier, just change some CSS
one of the funniest (and sadly accurate) things i’ve heard said about linux backwards-compatibility is that its most stable API is Win32. you can run really old windows software on wine because they support stuff even windows doesn’t anymore.
of course this is because the expectation is that you can just recompile old software to work on new systems, which is not really a thing on window.s
The most stable system is one that is out of support. No updates == No breakage! 😄
It’s very amusing to imagine devs carefully watching for an EOL/EOS date and starting to build software only after
I think this is the point of Debian stable, and also why some devs hate Debian stable.
It’s so portable! With maximal efforts we support both windows 7, windows 8.1 (but not 8.0), windiws 10 and soon Windows 11 !!!
/s
You can even do inefficient UIs in Python using tkinter, which is part of the standard library in python.
Python tkinter interfaces might be inefficient, slow and require labyrinthine code to set-up and use, but they make up for it by being breathtakingly ugly.
And now imagine yourself creating an UI in tkinter without an editor. Because that’s what I did. It was absolutely horrible.
Probably faster than me even deciding the bg color tho
I didnt care how it looked, because I only wanted for it to be finished fast.
For efficiency, just use the cli. The user doesn’t want it? Then I don’t want the user.
I do not want the user to ever be expected to edit the database by hand. If they make a mistake by doing so, they can break the entire system. That’s why I wrote a GUI which gets the job done.
Ugh, and they’re just weird. I can handle ugly but ktinker popups go across virtual desktops and over other windows for some ungodly reason, and never seem to dismiss themselves properly
Flatpak
AppImage
SnapHell, let’s not forget
Python Perl
Java
POSIXThe first 3 are Linux only. It’s irrelevant.
Also each is pretty bad in terms of usability and practicality, either losing integration because “containerized” or taking GBs of space or both.
Edit: guys relax, I’m not a linux hater, I use it daily. But windows does have a unified environment, which makes deployment so much easier, while linux doesn’t. And that’s a problem since you either have old broken apps on distro repositories, or impractical, potebtially bloated, and even more fractionated environments like those I mentioned. They are patches and we should work towards a more standard environment, not adding more and more levels of abstraction like electron does.
Even Torvalds says it so.
AppImages can get quite large because each app is self-contained, but the “losing integration” part is nonsense these days for any of these formars. That’s why we have portals, and if those aren’t enough you can still give the app full permissions.
Appimages are usually quite reasonable in size, it’s Flatpak that usually require 2/3 GB per app since every package has its own version of KDE/Gnome or other runtimes so every app still has to download a new one.
That’s simply not true, if the required dependencies are already downloaded they get used by every Flatpak app. If you have three apps requiring the Gnome 46 libs those only exist once.
I don’t know where this myth about Flatpaks always being gigabytes in size originates from or why it’s so persistent, but it’s wrong.
I don’t know where this myth about Flatpaks always being gigabytes in size originates from or why it’s so persistent, but it’s wrong.
Alright, here is ~25 GUI apps flatpak vs appimage:

- 6 GiB flatpak.
- 2.7 GIB appimage.
This is if you have a filesystem with transparent compression, If you do not have such filesystem (ext4), then it is more like 15 GIB vs 2.9 GiB lol.
This comparison is missing the flatpak equivalents of kdeconnect, deadbeef and a few CLI tools that I have on right btw,
flatpak-dedup-checkerfor some reason doesn’t check the/var/lib/flatpak/repodirectory which is usually another +1GiB in best case scenario.In my experience the issue is different flatpaks depending on different versions of gnome platform or mesa or whatever, so you have multiple versions of the same library.
I have plenty of storage though so I don’t care that much.
I’m not saying that’s not true.
I’m saying I’ve almost never downloaded a Flatpak that didn’t require a new dependency downloaded.
When I removed all my flatpk some time ago, I had: Steam, Viking, Discord, FreeCad and Flatseal to manage them. All of them and their dependencies used something arounx 17 GB of disk space (most of which was of course several versions of dependency runtimes), and that was after I removed all the unused runtimes that forn some reason it doesn’t remove after I uninstall or they are upgraded.
I’m sure if I installed more Flatpaks, some dependencies would eventually be reused, but you still need a good collection of them at any given time. So in pracrice you still need a lot lf space unfortunately.
losing integration because “containerized”
Bollocks. I’ve seen that many times with Flatpak (can’t speak for Snap), and every single time it was either because the packager failed to set up permissions or because the user messed with permissions that the application needed. Break off the tip of a screwdriver and it will no longer function as a screwdriver.
And I know you’re talking out of your ass because AppImage isn’t even sandboxed.
taking GBs of space
That part is true and accurate, and for a very good reason: dependency pinning. System packages can break if they don’t have the correct versions of shared libraries. If a package requires a very old version of a library, and doesn’t link it statically or supply it with the package, it can misbehave, have missing features, or refuse to even start. Flatpak (and probably Snap too, can’t speak for it) solves that by letting the packager specify (pin) the exact version of a dependency. If five separate packages require five different versions of the GNOME application framework, then they will download five separate packages of the correct version. AppImage solves it by being monolithic: everything is packaged together into a single executable.
losing integration because “containerized”Bollocks. I’ve seen that many times with Flatpak (can’t speak for Snap), and every single time it was either because the packager failed to set up permissions or because the user messed with permissions that the application needed. Break off the tip of a screwdriver and it will no longer function as a screwdriver.
Well then I guess you haven’t tried to get a password manager like KeepassXC to work with a Flatpak browser, because none of the solutions I’ve seen are “fix the permissions”.
From what I’ve read it seems you shouldn’t run a browser as a flatpak anyway, as this somehow weakens the built-in isolation.
I think I originally read about this somewhere librewolf-related but can’t seem to find it now.
I did find this similar discussion: https://discuss.privacyguides.net/t/correct-way-to-install-browsers-on-linux-securely/27046/6
I don’t know if it’s still the case, but up to a couple of years ago, Flatpak was configured so that externally mounted folders were not accessible. I discovered that when Steam on flatpak refused to install games on my hdd, and it was quite frustrating to figure out how to enable it. Still, it’s difficult to criticize how “bloated” are electron apps (they are) when I need to download 2GB or runtime for an 80MB telegram binary
Snaps integration is even worse as I’ve seen browser extensions state they straight don’t work on snap’s browsers. Also desktop integration on gnone (even files drag and drop between snaps) are broken on the ubuntu installations I tried.
Appimages have the least drawbacks and are my preferred methods between the three (at least they take less storage space than an equivalent Flarpak for some reason, but are still broken sometimes), yet they still miss a central package repository, and that’s a big problem.
yet they still miss a central package repository, and that’s a big problem.
It’s a package repository, but I would hardly call it “central”
I like having options
Flatpak is good. I say that as both a user of them and a dev of applications that are published as flatpaks.
.net
.net cli apps are cross platform and can be portable :p
Gui in .net isnt fully cross platform ( maui is everything except linux ) but frameworks like avalonia ( .net ) and imgui fix all that :')Does not MAUI compile to Linux, just that you need to compile on Windows?
Nope. The ui framework doesnt support linux at all. None of its cross platform claases and enums have linux. Its stupid because .net and aspcore just work on linux…
https://github.com/dotnet/maui
Android, iOS, iPadOS, macOS, and Windows
Everything but Linux, that’s funny.
I look at MAUI a year or so ago and IIRC, “official” Linux support was made by the community.
Thanks for reminding me why Maui doesn’t support Linux. I saw Maui mentioned in an earlier comment and was baffled why KDE would make something not working that doesn’t work in Linux. It’s because Microsoft stole the Maui name from KDE: https://www.phoronix.com/news/Microsoft-KDE-MAUI
LOL, didnt know maui naming was taken before. Hilarious
@DacoTaco @MoogleMaestro there is a port of mono winforms implementation to modern dotnet and it’s really works:
https://github.com/DanielVanNoord/System.Windows.FormsWinforms of all things haha. Eeuw xD
But thats flipping awesome and i love it! I know xamarin was wpf being cross platform with mono, but thats a dead horse thanks to microsoft haha@DacoTaco Yes, it’s Microsoft decided to kill classic win32 widgets and any XAML support in Mono, forcing new WinRT platform and not providing any GUI support in netcore. It’s sad that nobody made some crossplatform WPF implementation for modern dotnet, Maui is not incompatible with existing source code, IIRC
It’s sad that nobody made some crossplatform WPF implementation for modern dotnet
That’s basically Avalonia UI.
I believe Uno Platform is similar, but less of a WPF variant, and closer to Xamarin or MAUI in the style of Xaml that they use.
@hdsrob
>That’s basically Avalonia UI.
no, avalonia cannot help you run existing applications. Also. Avalonia is a crap, taking minutes to parse xml crap on poor hardware
Yes, there are many implementaion of WPF-like UI, but no implementation of existing Windows UI.
mono’s winforms allows you run existing winforms apps without single line change or even without recompiling (just replace related System.Windows.* assembies)
Avalon is actually xaml and cross platform. Personally id rather have a razor format of gui code, but xaml is a good second. Its basically what maui was suppose to be
Lazarus gang, this is your chance!
GTK is not accessible anywhere other than Linux and is therefore not a serious option outside of Linux.
GTK is available for windows. https://www.gtk.org/docs/installations/windows
It is. It’s just not particularly good outside of a X11/Wayland environment.
I think this being worked on though.
@jollyrogue @XenGi it is not particularly good outside of Gnome environment
Gtk2 and Gtk3 is fine outside of gnome. Gtk4 especially with Libadwaita is heavily focused on Gnome.
I said accessible not available. If you don’t know what it means, accessibility means integrating with the OS screen readers and other similar tools so that everyone can use your app, not only people by people with good eyesight, and capable of using a mouse.
I mean… Windows could probably contribute to GTK if they wanted to expose their accessibility APIs. This would in theory be a good use of time.
Thanks for the lesson
You need to put this in clear [JOKE] tags, otherwise they’ll never get it.
Every operating system contributed to the bloat. Windows has Win32, OS X has Carbon / Cocoa, Linux has X11 and various widget libs that sit on top of it. So it has been a perennial nut to crack to make cross platform widgets - wxWidgets, QT, SWT/JWT/Swing on Java, XMLShell (Firefox), Electron, GTK/GTK#, winelib etc.
Throw mobile platforms into the mix and it’s an unholy mess. Lowest common denominator is HTML and so the likes of Electron “wins” even though it’s bloated and slow.
i actually don’t have a problem with HTML, i just think that instead of every app shipping their own copy of electron, the operating system should provide basic browser functionality.
that sounds like Tauri!
Linux and Mac use WebkitGTK, Windows uses Edge/Chromium, Android uses Chrome - as bundled in the respective OS, and you essentially have a frontend running on that webview communicating with a backend running locally via some special IPC protocol
the operating system should provide basic browser functionalit
Microsoft got literally sued and almost dissolved as a company for that
But that’s cause it wasn’t FOSS, but instead a privately owned closed-source app
If an open standard was set, and agreed upon by most, then nobody would sue anyone
Heck, many Linux distros come with a browser preinstalled, but use a FOSS one to not hit that legal problem
Yeah, 1998 was definitely a time where Microsoft wasn’t going to do that. They were (and are) in embrace, extend, extinguish mode
If anything they’ve ramped up on the EEE mode
AzurEEE
And what did we get for it? If you search “chrome install” in Edge it pulls a Janet, all like

To be fair… Chrome is a lateral move on Windows. If I were forced to use Windows I’d probably just use edge if I needed a chromium browser lol
Then that operating system gets hit with anti-trust
Not if it’s FOSS
not if it’s a library
Abstraction layers? In MY messy pile of spaghetti ass code!?
You wouldn’t understand it, it’s more of a messy pile of lasagna ass-code situation.
It’s more likely than you think.
that’s a lot of words to basically say “i’m a fucking idiot”.




























