Is it really the best solution to create a fork?

Why not invest in people to directly work on Firefox?

The author even mentions a few paragraphs before that sentence that maintaining a browser is hard:

> The problem here is that making a capable browser is actually incredibly hard, as the browser has become a hugely capable platform all of its own. Undertaking the mammoth task of building a browser from scratch is not something a lot of people are interested in [...]

And who guarantees that the fork will further exist without weird funding?

EDIT: sorry, after reading my comment again, I think it sounds a bit snarky. English is not my native language, and I don't know how to phrase these questions better, so they don't sound that way.

Linux is not an operating system for the unwashed masses, it's a specialist operating system for more technical oriented folks. It's less polished in many areas and requires some jumping through hoops at times as we all know. In return, you get a vastly better user experience if you're the right kind of user.

Whenever I am placed in front of a standard Windows install, I'm shaking my head in astonishment and wonder how people put up with that. Yet, if I were a standard Windows user placed in front of a linux machine, I'd probably do the same.

We're just a different species of users. If linux users cared as much about browser specifities as they did about other things, I'm sure they would already be fixed. But the truth is: as a long-time linux user I can state that Firefox is just fine. Really, it is. For me. And probably for many other happy linux users too.

Does that mean if couldn't be improved? Of course not. But show me the software for which you couldn't say the same. Okay, maybe Emacs. Just kidding.

> The best and most visible example of that is hardware video acceleration.

That's not "Desktop Linux has a Firefox problem", that's "Desktop Linux has a problem". hardware acceleration was always a pain to get working in linux, as a subset of the larger problem of linux video drivers having varying levels of support for the plethora of features that video cards need to do these days.

Things like this ultimately led me to my current Linux usage: Windows host, linux vm, fullscreened on its own virtual desktop.

If I need to do something that linux wasn't going to do efficiently, I do it in the windows host. Otherwise, I do it in the linux vm.

Anyway, I wonder if any of the problems this person is pointing out are going to be made better or worse by wayland

Of course, we must be realistic. Linux around 6% is very different from MacOS at 6%. In that case, they already had support from items people complain to this day about Linux not having; main examĺes are msoffice and photoshop.

Nevertheless, Linux on the desktop has never been so good. Flatpak (or even AppImages and snaps) allow me to have a "stable core" with recently updated software. Support for most hardware is much better now (looks like most vendors make sure the hardware supports Linux, even though they themselves don't announce it). Pipewire and Wayland matured to the point where you can finally stream your desktop on Wayland using Pipewire. GNOME is snappier today then it was and cleaner and more elegant; of course we still need some consistency but windows suffers from that too; and, look, it even has thumbnails on the file picker! The kernel evolved to a point where desktop is just another well supported system: MGLRU and other improvements made latency on the desktop even under memory pressure just ideal.

The environment around it also evolved a lot. Many FLOSS software are on a quality level today that mostly only professional on almost niche areas can't use Linux on the desktop. Consider Blender, Godot, Audacity, Inkscape, Firefox, OBS... these things are refined, stable, elegant and don't try to steal your attention, require periodic payments or throw ads on your face. Actually a Linux desktop user feels very sorry when they see a "standard" windows user. And I'm not even talking about proton or areas where Linux leads or is well established.

So yes, it improved. Nevertheless, I'd love to, but I don't think I'll see Linux on the desktop beat windows or MacOS. But, know one thing: I don't care. Linux on the desktop has been good enough for me for a very long time and it will only get better. I just hope someday its market share will be big enough for it to no longer be ignored by so many vendors. Watching current growth, I don't think that is too far anymore now.

Google pays it to select it as the default browser, but Firefox' tiny declining market probably makes Google want to not to continue this except perhaps as a way to avoid monopoly stuff.

I'm considering writing letters to some of the sites that have the biggest issues and raising it as an issue.

Otherwise, Firefox is mostly fine. I have a few gripes with it and Moz are mismanaging their projects, but desktop Linux is hardly their #1 issue.

Or, is the problem in the complexity of the modern web browsing experience, and the walled garden created around it by current major browser developers (new standards, etc.)?

Hopefully then they’ll stop reinventing tabs, ruining privacy, and focus on the aforementioned bugs.

'Defining' Linux support:

>> Firefox is first and foremost a Windows browser, followed by macOS second, and Linux third. The love the Linux world has for Firefox is not reciprocated by Mozilla in the same way, and this shows in various places where issues fixed and addressed on the Windows side are ignored on the Linux side for years or longer.

80% of all revenue is from Google:

>> The giant sword of Damocles dangling above Firefox’ head are Mozilla’s really odd and lopsided revenue sources. As most of us are probably aware, Mozilla makes most of its money from a search deal with Google. Roughly 80% of Mozilla’s revenue comes from Google, who pays the browser maker to set Google Search as the default search engine.

Firefox declining usage and anti-trust risks with Google.

>> How long will this deal continue? Will it be renewed indefinitely, regardless of how much farther Firefox slides into irrelevance? Will the size of the deal drop, or will it end altogether? When will Google decide that spending hundreds of millions of dollars every year in what is essentially charity for a competitor is no longer worth it, or needed?

>> Google’s similar search deal with Apple is already facing legal scrutiny; will that scrutiny have consequences for the deal with Mozilla, too?

Solution:

>> "Linux needs a browser engine that is independent of Google (and Apple), and takes Linux seriously as a platform."

Mozilla could have done that by being a Rust consultancy which can fund the development of both Rust and Firefox without being heavily dependant on Google. That opportunity has been lost.

Realistically the likely outcome is to just break up Chrome from Google and use the Chrome engine as the standard browser engine, just like the Linux kernel is the standard kernel for all distros.

Case solved.

The Linux browser escaped out to conquer (konquer?) the world.

Konqueror was a nice browser, my fav back in the day, and it was exciting when Apple picked it to defeat IE with Safari.

However, maintaining a variant of Ungoogled Chromium or Vanadium targeted at desktop Linux would be feasible.

So what are you proposing then? It seems so irresponsible to complain about new-buzzword-shittification of something thousands of people contribute to and offer no contribution in return.

I remember Linus and Luke from Linus tech tips on YouTube making a series about the Linux desktop, in the end they both had a bunch of very constructive criticism, things that the "more technical oriented folk" would just brush off as minimal issues that can be dealt with it at the end of the backlog. But those issues are actually quite important, and if there's a chance that Linux can be open to more people, then let's do what we can to make that chance a reality.

Fans of the browser (starting with the guy in my mirror) need to start subscribing to services like their VPN or something to give Mozilla a better financial footing.

Media players like mpv supported VA-API for video acceleration for many years and that just worked for me without any issues on all my computers.

It is clearly a Firefox issue that it did not support API that was supported by many Linux-native software.

Windows 10 + WSL

I thought an AMD + Radeon laptop would be a great improvement. Evidently I have the Reverse Midas Touch. Also garbage. Sadly, FULL hardware/peripheral support requires Windows or Mac, happiness everywhere else dictates Linux, and WSL delivers both (plus WebGPU, which is another aspect of Firefox/Linux that is perplexing and half-baked).

On both computers, X was a dramatic improvement over Wayland: proper color support (i.e. parts of the GIMP UI wouldn't turn magenta or half-transparent, you could enable HDR in Chrome), no extreme keyboard lag, all my Java binaries work without requiring cryptic flags, I didn't have to compile my own drivers from a guy named ElFarto, I could finally forget the difference between AV1 and VP9 and whichever gstreamer package lets you watch video again...

I fret Everytime I update, yet I must update due to work. but I have to schedule and plan those updates so that when the whole thing takes a shit, I can spend 2 or 3 hours fixing it. Just using the built in update utility, or apt upgrade, or really anything runs about a 10-20% chance of shit breaking. (I didn't do the math)

audio experience is terrible all around, and Bluetooth and wifi use out the box, is meh at best for range and stability.

I've tried a dozen distros to see if the grass is greener somewhere else, it isn't, and I have to work, I don't have an hour every day to fiddle, I have to bill clients.

good support is almost non existent, IF you get help, it's from some greater than thou righteous asshole who suggests you just rewrite the drivers yourself and create a pull request.

while I agree it is close, there is a still a LOT of 'polish' that needs to come where I will feel confident that hitting update doesn't ruin my day, and that when I'm done with a stress filled day of work, I'll be able to boot up and play my games without having to troubleshoot for an hour first.

This has it at 5.1%. But ChromeOS has a lot of corporate mojo making it work. It’s only partially open source. A lot of the good bits are closed. So that puts Linux at 2.91%. A rounding error. For 30 years.

Hardware video acceleration is fine now, and apparently has been for some time.

But e.g. Google Meet plainly refuses to blur the video call background when run on Firefox. It happily does that when running in Chromium.

On ff I sync my add-ons, bookmarks and containers. What’s not to like?

Open a 1080p YouTube video and watch your processor get slammed. No, I do not want to install an unvetted third party plug-in. If nothing else, flag that the browser doesn’t support hardware acceleration when playing videos, in easy to understand terminology so I, and thousands of others don’t have to go troubleshoot exactly what is causing my computer to choke.