Chewy @ Chewy7324 @discuss.tchncs.de

Posts

96

Comments

1,011

Joined

2 yr. ago

I didn't know it was chosen not even 10 years ago, as it felt like it could've been around for longer.

tl;dr
Read the first sentence after each citation ;D

So Wayland, a protocol, is needing the addition of other protocols?

Yes. What we know as Wayland is the Wayland core protocol and a few other protocols that are absolutely necessary for desktop use (stable).

Then there is staging, which is not necessarily implemented by all compositors, e.g. fractional scaling.

Unstable also exists, which is even easier to get a protocol into (idk the exact requirements, likely the amount of support and explicit dislike by contributors).
These are often only used by a subset of compositors with e.g. XDG decoration allowing compositors to announce to clients (windows) that they support server side decorations (top bar with close/minimize/maximize buttons). This isn't implemented by Gnome, but most other desktops support it.

Different desktops also have their own protocols, which are published so that apps targeting those desktops can implement them. Some are also supported by other desktops, if they think they are suitable for them.
E.g. wlr layer shell makes status bars possible, which are used by basic compositors like Sway or Wayfire. KDE also supports it, even though it was originally created by wlroots.

Does this make it a Wayland Distro?

In a way you could say that a compositor is a Wayland distro, as it implements a subset of Wayland protocols.

In the end this is good, because it allows for rapid development and discontinuation of protocols. E.g. if a better protocol comes around, both protocols can be supported at the discretion of every compositor.

The goal was to solve the problem of X11, where Xorg still has to support drawing UI by itself, even though no program or toolkit uses it anymore (the 80s were very different). The Wayland core is so minimal there shouldn't be any issue with using it for a very long time.

Also, Wayland was developed by people with the goal to use it in automotive and other industry applications, where basic desktop functionalities, like multiple windows or session lock, aren't useful.

Cheat makers are likely behind this, asthey have monetary incentives to do so. If its Linux users I'd feel bad because stopping others from playing just because they can't, is extremely bad behaviour.

I like the Moz://a branding, altough most people wouldn't get it, so it makes sense to switch to correct spelling.

Whether the T-Rex is the coreect choice, is another question. I do like that it feels more creative than the basic, reduced logos of today.

Edit: I do like the new Logo. It looks good and it does match its "activist spirit". Mozilla the corporation is different from the foundation, and I do believe, that Mozilla is closer to its roots than all other browser vendors - including the reskins of Chromium.

Blender and DaVinci Resolve work better on Nvidia. AMD might work, but it will be a hassle and you'll likely need the proprietary AMD drivers anyway.

With Nvidia supporting Wayland and the open-source NVK continuing to get better, you could even switch to open source drivers for gaming at some point, if you prefer.

Edit: I've had enough issues with AMD GPU's clocking down while gaming, leading to micro stuttering. So don't buy AMD just because everyone tells you they work flawlessly.

For CPU and mainboard, everything works well — just don't buy a random unknown SSD from Amazon, then you're asking for data loss and random issues.

He said somewhere that he did ask a top contributor if they care, and they didn't. He also said that he rewrote a bunch of code to be able to change the license.

I can't verify this, but it doesn't seem like he infringend on someones copyright. Small changes (e.g. a few lines) don't even (necessarily) qualify for copyright (just like the few sentences I wrote here likely don't).

That's only true for new phones. Once they get older, the difference in performance is noticeable.

Replacing the battery after a few years is possible, upgrading the SoC is not.

Agreed. That's even better.

A default group name, which can be changed individually would be great.

Yes, but that argument can be made against TOTP too.

SMS 2FA is less secure than TOTP, but still better than no second factor, since most criminals with access to a password database aren't able to take over your phone number (on a large scale).

With most services allowing some kind of password reset over SMS, I also prefer no 2FA over SMS 2FA. I already lost an account because I changed my number.

With uBlock Origin and many other extensions working (almost) flawlessly on mobile, I'd say they were spot on. Almost, because some aren't comfortable to use because they aren't made with touch and small a screen in mind.

Yes, there's many ways to make programs unable to use other network interfaces. E.g. I'm creating a network namespace with a single wg0 interface, which I make services use through systemd NetworkNamespacePath.

That said, I'd argue gluetun is pretty much foolproof, especially with most people using docker which messes with iptables (edit: although I don't know if this'd be an issue for this use case).

Yes, and I seriously don't expect Ladybird to get anywhere near being a complete browser like Firefox.

Even the idea of being a "web standards first" browser seems prone to failure, looking at how many websites these days "work best on Google Chrome".

Firefox follows web standards pretty closely, and then some websites don't work correctly because they don't support a new Chrome feature not yet in a proper standard. How will this be different for Ladybird.

I'll be positively surprised if Ladybird gets to a point where it works for all websites, just like I hope Firefox continues to do the same.

I also think the Element Web UI is lacking, but it's gotten better over the last few years, after they started taking design more seriously. With Element X they do proper UI/UX design as a first step, and then implement it.

The old Riot.im client was exceptionally terrible, in performance and design, so I'm really happy with Element X.

Element being focused on corporate needs is nothing new, since they've a few large (government, healthcare) contracts, and they've struggled with financing for years now. Big deployments using Synapse is the big reason dendrite doesn't see much development anymore, even though it was planned as a replacement for Synapse at first.

I believe many of their side projects (P2P, VR) exist because they try to find possible business avenues, although I feel like most of them aren't successful (and they stretch to thin because of that).

Unencrypted messages are useful for very large rooms, where encryption doesn't provide meaningful more privacy since public rooms have to be considered public space anyway. Encryption does have overhead, so it makes sense to disable it.

Private rooms are E2EE by default and can't be created unencrypted (at least in the Element X mobile UI). This is a good way to handle it IMO.

Discord uses their own screen sharing implementation because it performs better than what's available in Electron by default. I don't expect Element to achieve that, considering their focus isn't gaming.

The user experience is generally worse than Discord, like any federated system compared to centralized platforms.

There is Cinny, a client with an UI similar to Discord. Element X is a great mobile client, and imo far superior to Discord for 1 on 1 chats (to be fair, I really dislike Discord 1 on 1 chat experience, so I'm biased).

Edit: It's worth noting that Element X does not support Spaces yet, which allows for grouping of rooms similar to Discord Server.

• Terminal file manager are useful on a server over ssh.
• ripgrep and fd support is better than any GUI file manager find and replace.
• Some people like using vim keybindings
• The three panel view is really useful. On the left is the parent folder, the middle the current and on the right a preview, e.g. the selected folder or the contents of a picture or a text file. It's faster to navigate and pop back into the shell.

What they mean with AI features is also their offline website translation feature, which is something I've wanted for years. The alternative is online Google website translation.

It's more so that you are allowed to try to block ads. If YouTube adds DRM or otherwise manages to stop usage of ad blocking, they are completely in their right to do so.

IIRC if something runs on your system, you have the right to mess with it, however you want — e.g. block specific parts of the site, or use reader mode etc.