"I'll have to check automod’s regex. We want to remove Discord invites as people do spam non-topical invites across many subreddits, but it should block based on URL, not the word."
The Chrome 91 branch, which (according to the article) the client is built upon, was released in May last year. That’s not great… …but not terrible either. I trust that, in due time, Discord will update to a newer version.
As for Flatpak… Fine by me, but don’t make it the only option, as suggested by this article. Some of us prefer having an up-to-date operating system with shims to support software that require older libraries, rather than the other way around.
I guess by EEE you mean GNOME is pushing for apps to use Flatpak, then once they do, pushing for the Flatpak version to be the main (or only) version. I don't think that really fits the EEE model because any Linux distro (or even Mac or Windows or BSD) can ship Flatpak.
Don't feel put upon, frankly they don't support Windows either. Their Electron app is so bad 'or maybe Electron is) that I'm using the web app instead. I actually looked up info on what it was made in that could peg my Core i5 at 99% when idle.
Couldn't agree more. Using the webapp with a chromium-based browser is such a better experience (working screen-share, hw-accel, etc.) I couldn't imagine going back.
And especially with chromiums --app flag (I just have a .desktop file that Execs `flatpak run com.github.Eloston.UngoogledChromium --app=https://discordapp.com/channels/@me --enable-features=UseOzonePlatform --ozone-platform=wayland`) it really feels like an "app" experience (exception being, that pressing on links in chat opens them in a new chromium window and not firefox).
Ripcord can cause your account to banned irrestoribly and prevent the phone number associated being used for verification for months. The maintainer is aware of this but does not warn users of this risk at all.
Using a third party client is considered self botting and against the TOS. This part of the TOS are enforced.
The worst part is that this is what many end-users want. We every so often get some strong pushback on the KMK (mechanical keyboard) Firmware project over my admittedly-dogmatic insistence on using Matrix (and otherwise trying to move further towards the Free Software Needs Free Tools [1] ideal world). Roughly paraphrased, "Well relatively few people have a Matrix account but a ton of folks are already on Discord, why do you make the Matrix only officially supported platform and have a flakey-and-maybe-five-8s-uptime bridge to Discord as an olive branch?" Because it's a GPL project with no corporate backing and no need to add dependence on another corporate tool (using GitHub for VCS and CI was, IMO with retrospect, a mistake too).
I've personally started taking a "if your project only offers me Slack and Discord as first-class avenues to get help, I consider it unsupported" stance on FOSS projects lately. It's bad enough that I have to use Slack for work and will never really have a say in that, don't make me also use it for my hobbies.
Maybe spending my adolescence in the era of IRC, of numerous IM protocols all accessible via libpurple+Pidgin/Adium, and of somewhat-open XMPP federation spoiled me. I clearly wasn't meant for this hyper-corporate modern flavor of the internet.
The interface of the popular matrix clients like Element is pretty good, certainly better than IRC, but a step down from Discord imo. I’ve submitted multiple bug reports for really basic stuff, like editing a message with a link in it (at the time the message could only be edited up until the link).
I’m sure it’s improved now but I don’t blame people for going to discord. If we want people to want to use open source software it would be nice if it was at least as good as commercial software
The great thing about Matrix over Discord is that you have the option of using or building a client that better suits your needs. With Discord, I'm stuck with the myriad of UX decisions they've made that I disagree with, unless I want to risk a ToS ban for self-botting (Ripcord, which I do generally enjoy).
I was permanently banned within a week of using Ripcord, with no other odd activity. I'm wondering if there are some lag times or picking a bad start point where Ripcord was slightly out of sync with newly released API calls or similar.
I had a longstanding account force me to verify with a phone number not long after I began using Ripcord (it wasn't banned, though, and I stopped using Ripcord after this happened)
Why do people insist on using Discord when Discord so clearly doesn't want them? There are alternatives (For example: Matrix: https://element.io/https://matrix.org/)
If someone is clearly and loudly telling you they don't want you and to go away, listen to them!
This is my whole point of 'defining Linux support' where the chronic inconsistency and mess that the Linux Desktop has created with competing system components conflict with each other and certain apps and this cannot be fixed.
Hence why you can't support all the Linux users, 100% of the time. Not even Discord can either. Since you get the complaining Linux user on an exotic distro with strange system-specific requests like supporting Wayland, XDG, Flakpak, GNOME, PipeWire, etc or not moving to Xorg which the devs have to hunt for the bugs all over the OS stack, but could end up only affecting that one user.
This is where developers draw the line and slap UNSUPPORTED on whatever exotic Linux distro you are using and 'letters' or 'requests' like this get ignored.
Due to that, this blog post explains why companies like Discord prioritise requests on predictable and consistent OSes, like Windows and Mac.
No where in my sentence did I mention the Steam Deck. My point spans beyond the Steam Deck and on to 'Linux Desktop' support in general.
Assuming you have read the article, it has already told us that Discord does not care to support the Linux Desktop more than Windows or Mac and would rather keep it outdated and working on most distros than to update everything and it ends up breaking on your favourite distro.
No company would seriously cater their resources to fully supporting an infinite about of distros to satisfy a small minority of non-paying users. Anyone expecting this is frankly trolling and know they cannot define Linux support themselves and writing such letters of 'free support' like this one is also futile.
I see nowhere in the article where it says fixing these issues (which largely just amounts to updating the version of electron they use) would cause it to break most distros.
Please point me to where it says that, as far as I can tell you're just making up your entire comment.
You see, I already said to you that 'defining Linux support' is unfixable. That is my whole point and it is spans beyond this topic where this 'chronic inconsistency' can only be found in the Linux desktop ecosystem.
> I see nowhere in the article where it says fixing these issues (which largely just amounts to updating the version of electron they use) would cause it to break most distros.
If it doesn't affect their favourite distro, then they won't care to write such letters or reports and that won't be their problem. It is Discord's or largely any software maker's problem to define this Linux support (which everyone knows they cannot.)
And no just 'updating the version of electron they use' isn't just as straightforward as the author affected by this issue would want to believe; especially with the tons of Linux distros some users out there and its ecosystem of alternatives.
> Please point me to where it says that, as far as I can tell you're just making up your entire comment.
Creating a strawman to attempt to refute my whole evergreen point about the Linux Desktop's issue (which is beyond this article) isn't a good way to convince me otherwise, unless Discord is a simple hello world app. If that was the case, this letter would not need to be written.
Unfortunately, Discord is quite more than that and its seems they are very hesitant to offer updates that can break the app on Linux.
Sure, but in this case specifically, "supporting (all) linux users" is already "solved" (they really only have to update their dependencies).
Also, nobody expects them to support "all" users. They are only supposed to update their dependency electron (which does the linux compatibility stuff already) and maybe maintain the distribution for one of their new gamer target groups (steam os).
> Sure, but in this case specifically, "supporting (all) linux users" is already "solved" (they really only have to update their dependencies).
And how do you know that it will 'just work' after upgrading their dependencies? My point spans more generally that is unfixable due to the level of fragmentation that is in the Linux Desktop ecosystem.
The fact that Discord hasn't done it tells us that they are both aren't confident in supporting an unstable desktop environment with too many moving parts or there is a low amount of users there which doesn't justify the urgency to fix these issues and will let it sit on a working version of Electron before they upgrade it and it breaks something.
> Also, nobody expects them to support "all" users.
Hence why, Discord knows that they will always get a plethora of annoyed 'Linux' users on an infinite amount of distro combinations reporting that the app doesn't work on someone's favourite distro.
Hardly the case for Windows and macOS users meeting the minimum requirements to run Discord where there is a guarantee of having 100% support all the time.
> I would also like to ask Discord to maintain the Discord Flatpak. Since the Steam Deck uses Flathub by default, having official support on Flathub would be a great huge step for Discord and the Linux community.
Hah, every single one of these "why doesn't $PROGRAM run on my system???" articles seem to be the same Trojan horse for Flatpak and Wayland support. Here's a thought: if sandboxing and stability matter so much to you, why not use the Discord Snap?
I'm really sick and tired of this. The GNOME maintainers keep shooting themselves in the foot with insane requests ("everyone, please switch to our New Competing Standard, Flatpak!"), head-scratching regressions ("moving forward, we will not be supporting themes/shell extensions/legacy display servers/AppIndicators/your hardware/etc.") and worst of all, development cycles so slow they could bore a sloth. You want people to get onboard with your tech? Build something. Coupling technology like Bubblewrap to Flatpak or Mutter and Adwaita to GNOME is what makes people cautious of supporting these kinds of transitions.
The real reason Discord sucks on this person's desktop is because neither Wayland nor Flatpak is the future. There is no "transition" to these technologies. GNOME is being different for the sake of being different, and complaining when everyone else doesn't rally behind them.
Sandboxing is important for the desktop security. The choices are between Flatpak and Snap for desktop Linux.
Snap requires lock-in to Canonical's proprietary cloud app store, whereas Flatpak you can link to multiple app stores simultaneously. It's obvious which standard is best for consumers.
I agree. Flatpak is a packaging technology though, not a sandboxing one: if privacy/security was the primary concern, then Bubblewrap would be shipped as a separate program for any software to run in, not just Flatpaks.
Flatpak I can see, but Wayland? I've been under the impression that X is going the way of the dodo in any case (maintainers becoming scarce etc.), and that Wayland has been (and is) the only real alternative that is being worked on.
Now, this is just my (very much non-expert) understanding on the matter, so if you have anything more substantial to add (or link me to) for further reading it would be much appreciated.
I mean, see for yourself: only 3 DEs actually support Wayland, the rest are WONTFIX or have too much to port to wlroots. GNOME Mutter has the only complete Wayland implimentation anyways, if 10 years go by and 1 DE has adopted you as default, you've probably failed.
