Quoting: no_information_here

Quoting: pleasereadthemanualA member of the Quality Assurance team for instance is concerned about being able to assure the quality of KDE:

QuoteFor KDE, this would be extra 51 apps! Compare the scope. I believe it's impossible for us to have the same quality bar here. And even if we had a huge surge of volunteers to test those regularly, it would just mean that we'd hardly ever release, because the likelihood of discovering a broken functionality in 73 apps (22+51) is much higher than in 22 apps. Workstation is quite lean on pre-installed apps, and yet we already struggle with this, and many people get irked by the whole compose being blocked on a bug in gnome-clocks/gnome-contacts/gnome-calendar/etc.

Thanks for the quote. I am not a Fedora user -- what do they mean by "apps" here? Other than the core plasma bits (settings, kwallet, etc.) the only "KDE apps" I have installed on my Neon system are Okular, Dolphin, and Gwenview.

Edit: I forgot Kate.

There are quite a lot of them: https://apps.kde.org

I don't know which ones Fedora's KDE SIG thinks are core.

View this comment - View article - View full comments

Quoting: LanzI see this as Fedora admitting that Gnome is a bit stalled at the moment.

There are a lot of Fedora users and KDE maintainers pushing for KDE to be better represented. Fedora's KDE spin (now Edition) has a lot more maintainers than it had in the past, so it's as well maintained as Fedora Workstation (GNOME), as compared to other spins like i3, Sway, and XFCE.

This isn't the singular opinion of "The Fedora Project" but like all FESCO decisions, a community of disparate people coming together to discuss this. Some don't have any opinion on KDE because they've never used it. The KDE and GNOME people are completely different aside from Neal Gompa. A quote from the issue tracker:

Quote(Honestly, the only thing we have in common right now is Neal imo, otherwise we don'T really communicate as-is)

A member of the Quality Assurance team for instance is concerned about being able to assure the quality of KDE:

QuoteFor KDE, this would be extra 51 apps! Compare the scope. I believe it's impossible for us to have the same quality bar here. And even if we had a huge surge of volunteers to test those regularly, it would just mean that we'd hardly ever release, because the likelihood of discovering a broken functionality in 73 apps (22+51) is much higher than in 22 apps. Workstation is quite lean on pre-installed apps, and yet we already struggle with this, and many people get irked by the whole compose being blocked on a bug in gnome-clocks/gnome-contacts/gnome-calendar/etc.

And concluding:

QuoteEither the quality requirements won't be the same, or we need to lower the Workstation one and meet somewhere in the middle for both.

This discussion has been going on for several months and it's taken this long because the KDE SIG are the ones interested in becoming an Edition, while the GNOME people mostly don't have an opinion. There were a lot of objections in the beginning about how confusing it would be to have two Editions listed equally. It was hard to see how it could work. Especially because no one wanted to get rid of GNOME because Fedora has a great relationship with GNOME.

So while it is an acknowledgement of how good a job the KDE SIG is doing at maintaining the KDE spin (despite some decisions not being in line with upstream like dropping X11 packages), this says little about what Fedora thinks of GNOME.

View this comment - View article - View full comments

Quote"Manjaro Data Donor", which they plan to have as opt-out and not opt-in.

I think this is fine if they give you an undismissable prompt telling you they're going to do it and letting you opt out.

QuoteI know telemetry is a contentious subject, but we need at least some data about how Manjaro is being used by so many people around the world in order to show that the project has a future and also to plan for that future.

I sympathise with the plight (even though it's Manjaro's plight), but I can't reconcile the morality of opt-out telemetry. If you do opt-out telemetry, you believe you are entitled to your user's data. Is there any scenario where that isn't true?

Another way of saying the same thing is:

QuoteIf we do opt-out telemetry, we'll snag a high percentage of people who don't even realize we're collecting data on them who would otherwise not consent to having their data collected if we made it opt-in. But this data is really important for us; more important than your consent.

I can't reconcile it.

View this comment - View article - View full comments

Quoting: StellaThat's cool, but i'm wondering how much 'older' does a gpu need to be to not support Vulkan 1.3?

Kepler and earlier don't support Vulkan 1.3: https://en.wikipedia.org/wiki/Kepler_(microarchitecture)

These are cards from around 2014 and earlier. Maxwell cards, which started releasing in 2014, support Vulkan 1.3.

For AMD, it seems to be from GCN 3.0 and earlier; GCN 4.0+ supports Vulkan 1.3. In AMD's case, it's cards from around 2015 and earlier that are stuck on Vulkan 1.2.

View this comment - View article - View full comments

Quoting: Purple Library Guy

Quoting: Linux_Rocks

QuoteBetter Dvorak keyboard detection.

The Dvorak keyboard users right now:

To protect your privacy, external media requires approval to load. Source: i.giphy.com
View cookie preferences.
Accept & Show Accept All & Don't show this again   Direct Link

I have come to the conclusion that there used to be a point to the Dvorak keyboard, but there mostly isn't any more. The thing is, the Dvorak keyboard lets you type faster because the layout is more efficient given the frequency with which letters are used and stuff. But nobody needs to type fast any more. Keyboards are ubiquitous, but nobody is using typing to copy text or take dictation. Instead of high speed data entry there's barcodes, QR codes, and copy/paste because the information was on computer already in the first place.

So people are only typing stuff as they think it up, and most people can type on a QWERTY keyboard as fast as they can compose. Far as I can tell, people mostly don't even bother learning to touch type any more. So Dvorak becomes a case of "solving a problem people don't have".

As someone who used Typeracer every day for several years (and TyprX more recently..), I have never seen the value in DVORAK. It would take too long to re-learn where the keys are and get to a speed that even comes close to how fast I am with QWERTY.

It would be far more productive to learn stenography with software like Plover. It might take some time to learn, but you will reach speeds much higher than you ever could with QWERTY or even DVORAK.

But 80WPM is already far faster than people can think, and that's probably an average typing speed for a professional computer user. I once had the immense displeasure of transcribing a 100,000-word faded (and insane) manuscript from the '70s, which is the closest I ever came to a practical professional use for typing quickly. It was so hard to read, though, and so I didn't end up typing it that quickly... in fact it was because the OCR had such a hard time of it that I ended up transcribing it myself.

Typing quickly is a useless skill today unless you're a court reporter. The only thing you're achieving is inflating your ego. But 100 years ago, typists earned decent money from winning typing competitions. Typewriter companies wanted to advertise their typewriters. I learned about this thanks to Sean Wrona's typing history video, which is a great watch...for people interested in typing: https://www.youtube.com/watch?v=uBL0J57HbIc

There are still typing competitions today, but you're not going to make a living doing it. And Sean Wrona was never a DVORAK user anyway :)

View this comment - View article - View full comments

Given past trends, "DayZ works really nicely on Linux for now" might prove to be more accurate.

I worry about getting into any Anti-cheat game with my friends because when they drop support for Proton, that will just become one more thing keeping me on Windows. And then I remind myself this way of thinking is dumb, because it doesn't matter how many reasons I have—one is enough.

And then there's the worry of false banwaves like Apex Legends had, I suppose.

View this comment - View article - View full comments

(though to be clear my opinion is Fedora should stop blocking the auto-download of openH264 in Firefox as I believe it results in a significantly worse experience for a benefit even free software evangelists would argue is dubious at best)

View this comment - View article - View full comments

Quoting: MothWavesjust feels so disingenuous with the way it tries to minimize the importance of their Linux playerbase and the way they refer to Linux like it is itself the problem.

Linux itself is the problem. It's not compatible with the most advanced modes of the most advanced anti-cheat software. Server-side anti-cheat is workable in some situations but does not work as well as client-side anti-cheat—else you might find more developers using Valve's server-side anti-cheat instead of all of the kernel level options. Competitive Counterstrike forces the use of FACEIT, a kernel-level anti-cheat, instead of VAC because it is more effective.

In fact, Apex Legends could have used VAC as it is a Source game, but they decided not to.

This was going to be a problem from the time BattlEye and EAC announced support for Linux and it was why they left it up to the developers to enable. I don't think there's anything more to be done. Competitive multiplayer is dead on Linux.

The only thing that will work, as Apex Legends says, is for Linux users to be such a huge audience that blocking them altogether would result in a noticeable drop in the amount of players. For now, it's a lot easier to block Linux than deal with the problems it brings.

I'm not hopeful but I would enjoy being proven wrong.

View this comment - View article - View full comments

Quoting: RalimbaTime to boycott those kernel anti-cheat games.

We can't play them anyway, so I guess we're ahead of the curve...

View this comment - View article - View full comments

Quoting: fenglengshun

Quoting: dziadulewiczIndeed. It's an absolute shitshow. It is mandatory to use terminal and somehow *know* commands and what else. You are expected to just *know* there is this thing called RPMFusion (that you have to manually enable, and from where to start with, also a mystery as not any website or link is given). The installer doesn't express any of this. You also need to install additional codecs.

Fedora doesn't ship patented media codecs by default as for example Ubuntu and Linux Mint do.

It is beyond any normally thinking user *why there are no couple of simple boxes to tick* during install to achieve this totally basic functionality to watch videos.

So basically a new comer "can't watch YouTube, Dlive and Twitch on Linux" OOTB if you happen to choose Fedora as a first distro.

It's codecs. It's related to patents. Patent laws are a mess. Ubuntu, and its downstreams, choose to ignore the issue. Fedora follows a stricter guidelines. There's really not much you can do with what Fedora's rules vs Patent rules.

If it doesn't fit your preference, then use one of its downstreams like Nobara, Aurora, Bazzite, UltrarisiOS, or risiOS which are more newcomer-friendly. Fedora is meant to be a very unopinionated and cleanly-operated community project

(H.264 refers to the video codec commonly used for the mp4 container, which I will be saying a lot)

Specifically, Fedora would love to include a H.264 decoder by default. Unfortunately, "Fedora Legal"[1] has determined it is illegal to include a H.264 decoder from ffmpeg, for example, unless they pay for it for every user of Fedora. This would cost many millions of dollars a year.

There is openH264, which Cisco created and negotiated with MPEG-LA (now Via-La) for an unlimited license. The condition was that every consumer of openH264 needed to use the binaries Cisco provided. Despite the project being open source, you are not allowed to make any modifications to it. By default, Firefox includes openH264 as an automatic download after the user opens Firefox. Fedora has intentionally intervened and disabled the auto-download, which you need to download a package to override.

This is because openH264 cannot be built from source without violating the terms of using it royalty-free. Fedora only includes open source software in its default repositories, and openH264 does not qualify with terms like these. Even if they aren't terms regarding the use of the software itself.

openH264 is inferior to ffmpeg's x264 in several ways. Even if it was included by default, it still cannot play H.264 High 10 profile media, so it may not play all of your local files. For the web, it is adequate.

In 2031, all of the current H.264 patents will have expired and it will be legal for Fedora to include a H.264 decoder and encoder by default. Fedora will still not include a H.265 encoder/decoder (to say nothing of H.266), but these formats fall back to H.264 99.9% of the time anyway. Fedora also does not include a VC-1 decoder, though this is a far nicher codec that I have yet to encounter in the wild. Maybe in some old visual novels..? The patents for this codec will likely expire before H.264; there's only a few left: https://www.via-la.com/licensing-2/vc-1/vc-1-patent-list/

Aside from this, Fedora currently includes an AAC decoder which would have been a point of contention in the years previous.

YouTube, Netflix, Facebook, and other media sites will fallback to AV1 or VP9 if they detect a H.264 decoder is not available, so if you only visit a few of these sites, you won't notice you're missing the codec. If you use Peertube, you're out of luck.

On Fedora Silverblue, this limitation will be a lot easier to solve. In the future, there will be a specially-prepared Flatpak extension in the Fedora Flatpaks repository that enables the openH264 in Firefox. You will be able to install it via GNOME Software. No terminal commands required. Still a pain, but far less of one. The real issue is most people don't even understand why they can't playback videos on certain sites or locally.

It's not as if Fedora does this to intentionally give people a bad experience. They're working within the limitations they have. Fedora team members have expressed a desire to change this, but what can they do, legally? Not much.

Further reading: https://wiki.endsoftwarepatents.org/wiki/H.264

[1]: "Fedora Legal" refers to Fedora's legal department, which doesn't technically exist. Some lawyers who work for Red Hat donate some of their time to answer legal questions Fedora developers have. That time is finite, however, and not all questions are answered in a timely manner (or at all).

View this comment - View article - View full comments