Quoting: GuestWell i think on most distros you can enable automatic upgrades, that's what i usually do on the stable variants, as they tend not to break (ever) with upgrades.

It's good to hear this. Fedora has an auto-upgrade option or plug-in through dnf. Flatpak already automatically updates which is a recent change I like. I still manually upgrade dnf repos through the terminal because I like seeing seeing what is being updated and changed. It's good to see other distros adopting auto-update behavior too.

The future, at least for mainstream OS's, will be immutable system images in a containerized app environment, like Fedora Silverblue, CoreOS, and others. That's a short ways down the road though.

View this comment - View article - View full comments

Quoting: PatolaThat trend of doing userspace/out of kernel drivers for devices (new-lg4ff, xow, sc-controller) should get some centralization. It could quickly go out of hand and besides being out-to-date with each other, could let some stuff lie in obscurity.

Would you mind expanding a little on how things could get out of hand and how centralization might work? This is something I don't know much about but sounds like it is important.

View this comment - View article - View full comments

SpaceTrains does look completely absurd and fun. I'll keep an eye on it.

View this comment - View article - View full comments

Crayta looks pretty cool and the platform really needs more titles in the library. The "state share" feature sounds like an answer to a manufactured problem. We'll see if Crayta sells it or not, but the game looks interesting enough to me without that feature.

View this comment - View article - View full comments

Quoting: Tuxee

Quoting: Pangaea

Quoting: TuxeeReally? Showing an image of a flatpak package to prove the bloatedness of a snap? Really? (Besides I can't find a XNView snap to see whether this is different.)

Maybe read what I posted then, instead of barking up wrong trees. With bloat I always talked about the hilariously ill-named flatpaks. I have never used Snaps and don't intend to, so have no idea if these are bloated too or if the downsides are more about the Canonical lock-in.

Sorry. Yes you were mentioning flatpaks. (From my experience snaps are not exactly bloated.)

The "XNView MP" flatpak on Fedora (through Flathub) is only 191.1MB (install size) on disk for me. Installing apps across DEs invovles a lot of overhead in every case. All the missing libraries and frameworks need installed. That will happen on Flatpak, Snaps, AppImage, or your distro package mangler. If you're running Gnome and install a KDE app that has a lot of dependencies those will get installed no matter the source system. What it looks like is that person needs a lot of stuff that isn't on their system already. With Flatpaks once that runtime environment is installed all apps that use that version will share it. Although I believe they must be from the same flatpak repo so Fedora and Flathub runtimes aren't shared since their repo sources are different, or at least that has been my experience installing GIMP, Inkscape, and a couple other apps.

View this comment - View article - View full comments

Quoting: MohandevirWhy not sell these Steam keys on the Epic store then? :D
Don't answer to that, just being sarcastic.

Your joke brings up a great point though. Why aren't game and media purchases recognized across store fronts, especially for service oriented games? The ecosystem is currently very user hostile and pro-publisher/distributor. If we buy a game from a publisher it should be considered 'valid and purchased' across many store fronts. If anything that will truly drive consumer prices down and provide incentive for distributors to make better deals with studios and publishers.

View this comment - View article - View full comments

Quoting: tuubi"You're not fans of a thing I like so you must be ignorant and get all your info on social media." With such a confrontative attitude, I'm sure you feel right at home on Reddit and YouTube.

Please accept that snap does not enjoy the kind of support from the greater Linux community that other controversial projects like systemd (or directly competing projects like flatpak) do. It's very much a Canonical project, and they seem to want to keep it that way. I personally don't mind at all that snaps exist as an option, but I simultaneously don't see why I should mind that Mint is not jumping on that particular train at this point.

Regardless of how useful container tech can be in many areas of computing, there are valid reasons why perfectly sane and knowledgeable people might not be enthusiastic about a particular use case or implementation. And many of those people have embraced projects like systemd, simply because they're useful to them in some tangible way. I'd claim to be one of those people, but I don't think I'm qualified to assess my own sanity.

Agreed. I don't personally see the broader community (outside of Ubuntu fans) as particularly enthusiastic about Snaps.

I'm not on Mint anymore so I don't have a direct stake in this. I have given Snaps a shot, just like AppImage, Flatpak, various distro package managers, PPAs, and the Almighty AUR. I didn't like the system when I was on Ubuntu or Manjaro. I ended up removing it on Manjaro and just using the AUR and Flatpak.

Snaps are proprietary, locked into Canonical, and in my personal opinion don't see the system as community friendly long term. It is a store system designed for single vendor lock-in which I don't like. All systems have their trade-offs and technical issues and boons. I'm not singling out snaps on that. Some of its annoying flaws are shared with Flatpaks which is a system I do like. The reason I can't support Snaps, as an option, is due to that lock-in and what I see as community hostile approach.

View this comment - View article - View full comments

Quoting: Shmerl

Quoting: randylFrom an analytics perspective, I don't feel it is misleading at all. Misleading would tell something that isn't. It's incomplete, not misleading. Reasonable conclusions can be drawn from the data such as a minimum floor of specific users. For example we could know that 10% of all users are Linux based and of those at least 47% are Ubuntu based. That doesn't mean there aren't more, so the conclusion we can draw is that a FLOOR value of people use Ubuntu. We can't draw a conclusion that there are exactly 'X' number of users, but that fuzzy set is still useful. It can be added to other sets of data as a point of reference too.

Sure, if you analyze it correctly, you can draw proper conclusions. Misleading means that it's easy to draw incorrect conclusions if you treat it as "look, Ubuntu must have this percentage". Which I easily can see happening.

I can see that happening too, but most data can be misinterpreted if a user isn't trained or has an agenda. I do agree that presenting distro data too strongly could easily be misinterpreted. In context of the data collection being able to assert claims based on the distro data has some pretty strong limitations.

View this comment - View article - View full comments

Quoting: Mal

Quoting: amataiThe 30% is only for game bought on steam. Valve gives dev key generator so they can sell how much keys they want without the need to pay a cut to Valve.

It's true also that to sell those keys you have to pay for a variety of infrastructure and services unless you want to drown in frauds. I came to realize that for small indies generating steam keys (outside the free ones for press and PR) can often result being a bad idea. Sometimes even when they sell those on sanctioned stores (see g2a controversy). But then they'll have to pay a cut anyway so...

Ofc none of this is Valve fault. Just nothing that choosing to renounce some of their services to save some bucks still mean that you have to pay to build those services or pay someone to provide them to you in their place.

Your answer is hyperbole in an effort to dismiss a valid claim. How does a studio generating its own keys suffer from fraud? Do you have any proof of that? What are the additional costs for self-hosting keygen?

A studio may suffer from stolen credit card fraud, but there is no inherent massive cost to hosting your own keygen. Many studios do offer keys through their own site and they don't have to pay the 30%. Authorized reseller sites also don't suffer from keygen fraud (again credit card, but not keygen).

I really need to see proof of your claims because they don't sound grounded in reality at all.

View this comment - View article - View full comments

Quoting: Shmerl

Quoting: KuJoIt depends on the browser used and from which sources it was installed.

Which means this method is pretty useless if not even misleading for analyzing distros data.

From an analytics perspective, I don't feel it is misleading at all. Misleading would tell something that isn't. It's incomplete, not misleading. Reasonable conclusions can be drawn from the data such as a minimum floor of specific users. For example we could know that 10% of all users are Linux based and of those at least 47% are Ubuntu based. That doesn't mean there aren't more, so the conclusion we can draw is that a FLOOR value of people use Ubuntu. We can't draw a conclusion that there are exactly 'X' number of users, but that fuzzy set is still useful. It can be added to other sets of data as a point of reference too.

View this comment - View article - View full comments