@Max_P@lemmy.max-p.me avatar

Max_P

@Max_P@lemmy.max-p.me

Just some Internet guy

He/him/them šŸ³ļøā€šŸŒˆ

This profile is from a federated server and may be incomplete. Browse more on the original instance.

Max_P,
@Max_P@lemmy.max-p.me avatar

I must be lucky, works just fine for me with SDDM configured for Wayland only, autologin to a Wayland session.


<span style="color:#323232;">max-p@media ~ % cat /etc/sddm.conf
</span><span style="color:#323232;">[Autologin]
</span><span style="color:#323232;">User=max-p
</span><span style="color:#323232;">Session=plasma
</span><span style="color:#323232;">#Session=plasma-bigscreen
</span><span style="color:#323232;">Relogin=true
</span><span style="color:#323232;">
</span><span style="color:#323232;">[General]
</span><span style="color:#323232;">DisplayServer=wayland
</span>
Max_P,
@Max_P@lemmy.max-p.me avatar

Arch. That leads me to believe itā€™s possibly a configuration issue. Mine is pretty barebones, itā€™s literally just that one file.

AFAIK the ones in sddm.conf.d are for useful because the GUI can focus on just one file without nuking other userā€™s configurations. But they all get loaded so it shouldnā€™t matter.

The linked bug report seems to blame PAM modules, kwallet in particular which I donā€™t think Iā€™ve got configured for unlock at login since thereā€™s no password to that account in the first place.

Max_P,
@Max_P@lemmy.max-p.me avatar

ActivityPub makes this impossible. Everything on the fediverse is completely public, including votes, subscriptions and usernames. Even if Lemmy did offer the option, other servers wouldnā€™t necessarily.

And honestly this is a system that would be mainly used for spam and hate speech anyway. Just make a throwaway like everywhere else.

Blame Canada? Justin Trudeau Creates Blueprint for Dystopia in Horrific Speech Bill (www.racket.news)

I got in some hot water a while back for admitting I was relatively unconcerned with Republican villainy these days compared to other worries. This Canada Online Harms Act, whose details I missed earlier (apologies to Public and Yuri Bezmenov!), perfectly embodies the kind of thing that keeps me up at night now....

Max_P,
@Max_P@lemmy.max-p.me avatar

Genuine political dissent would become logistically impossible, and virtual mob rule a certainty.

Thereā€™s a major difference between political dissent and hate speech.

You can say: ā€œI donā€™t think transgender people should be allowed to choose which bathroom they go toā€, that just makes you a shitty person with no compassion. But if you say ā€œIf I see a trans women in the mens bathroom Iā€™ll beat them up out of thereā€, that is very clearly hate speech and threats of violence.

Itā€™s not like itā€™s hard to treat people with basic respect. You can disagree without resorting to hateful comments and threats and name calling. If you donā€™t see the hate speech problem youā€™re probably part of the problem because surprise, itā€™s only conservatives you see online constantly spewing FUD and hate speech.

Max_P,
@Max_P@lemmy.max-p.me avatar

Have they tried building average, affordable condos instead of luxury condos?

All the new condo developments Iā€™ve been seeing everywhere are more expensive than a townhouse nearby, and half the size of nearby apartments that are also half what the mortgage would be on a 1-2M fucking condo.

Pretty sure they can do away without the penthouse pool and hot tub and bring the price down, but it wouldnā€™t be nearly as profitable for them so they wonā€™t.

Nobody is actually trying to help lower home prices because everyoneā€™s retirement plan depends on houses being unaffordable.

Max_P,
@Max_P@lemmy.max-p.me avatar

That, and they also either renovate in pseudo-luxery or build fairly luxurious condos in a world of HOAs going absolutely out of control and making sure you canā€™t actually live in there even if you could afford it because your next door karen will file a noise complaint every time you flush the toilet.

Number of constructions Iā€™ve seen actually designed to be affordable in the last couple years: zero. None. They all target rich people that could afford a normal house anyway.

Even new apartment buildings, the thing that people that canā€™t get a mortgage get, is now also all designed for 3000+ monthly rents.

Property developers are greedy pigs.

Max_P,
@Max_P@lemmy.max-p.me avatar

Fairly new to ham, whatā€™s nice to listen to during an aurora? Just funny noise bursts? Any antenna precautions so I donā€™t fry my SDR?

Max_P,
@Max_P@lemmy.max-p.me avatar

I route through my server or my home router when using public WiFi and stuff. I donā€™t care too much about the privacy aspect, my real identity is attached to my server and domain anyway. I even have rDNS configured, thereā€™s no hiding who the IP belongs to.

That said, server providers are much less likely to analyze your traffic because thatā€™d be a big no-no for a lot of companies using those servers. And of course any given request may actually be from any of Lemmy, Mastodon, IRC bots or Matrix, so pings to weird sites can result entirely from someone posting that link somewhere.

And it does have the advantage that if you try to DDoS that IP youā€™ll be very unsuccessful.

Max_P,
@Max_P@lemmy.max-p.me avatar

Iā€™ve actually ran into some of those problems. If you run sudo su --login someuser, itā€™s still part of your userā€™s process group and session. With run0 that would actually give you a shell equivalent to as if you logged in locally, and manage user units, all the PAM modules.

systemd-run can do a lot of stuff, basically anything you can possibly do in a systemd unit, which is basically every property you can set on a process. Processor affinity, memory limits, cgroups, capabilities, NUMA node binding, namespaces, everything.

Iā€™m not sure I would adopt run0 as my goto since if D-Bus is hosed youā€™re really locked out and stuck. But itā€™s got its uses, and itā€™s just a symlink, itā€™s basically free so its existence is kBs of bloat at most. Thereā€™s always good ol su when youā€™re really stuck.

Max_P,
@Max_P@lemmy.max-p.me avatar

Basically, the SUID bit makes a program get the permissions of the owner when executed. If you set /bin/bash as SUID, suddenly every bash shell would be a root shell, kind of. Processes on Linux have a real user ID, an effective user ID, and also a saved user ID that can be used to temporarily drop privileges and gain them back again later.

So tools like sudo and doas use this mechanism to temporarily become root, then run checks to make sure youā€™re allowed to use sudo, then run your command. But that process is still in your userā€™s session and process group, and youā€™re still its real user ID. If anything goes wrong between sudo being root and checking permissions, that can lead to a root shell when you werenā€™t supposed to, and you have a root exploit. Sudo is entirely responsible for cleaning the environment before launching the child process so that itā€™s safe.

Run0/systemd-run acts more like an API client. The client, running as your user, asks systemd to create a process and give you its inputs and outputs, which then creates it on your behalf on a clean process tree completely separate from your user sessionā€™s process tree and group. The client never ever gets permissions, never has to check for the permissions, itā€™s systemd that does over D-Bus through PolKit which are both isolated and unprivileged services. So thereā€™s no dangerous code running anywhere to exploit to gain privileges. And it makes run0 very non-special and boring in the process, it really does practically nothing. Want to make your own in Python? You can, safely and quite easily. Any app can easily integrate sudo functionnality fairly safely, and itā€™ll even trigger the DEā€™s elevated permission prompt, which is a separate process so you can grant sudo access to an app without it being able to know about your password.

Run0 takes care of interpreting what you want to do, D-Bus passes the message around, PolKit adds its stamp of approval to it, systemd takes care of spawning of the process and only the spawning of the process. Every bit does its job in isolation from the others so itā€™s hard to exploit.

Max_P,
@Max_P@lemmy.max-p.me avatar

I havenā€™t had D-Bus problems in quite a while but actually run0 should help with some of those issues. Like, systemctl --user will actually work when used with run0, or at least systemd-run can.

Havenā€™t used it yet so itā€™s all theoretical, but it makes sense to me especially at work. Iā€™ve used systemd-run to run processes in very precise contexts, itā€™s worth using even if just to smush together schedtool, numactl, nice, taskset and sudo in one command and one syntax. Anything a systemd unit can do, systemd-run and run0 can do as well.

Iā€™m definitely going to keep su around just in case because I will break it the same Iā€™ve broken sudo a few times, but I might give it a shot and see if itā€™s any good just for funsies.

Just trying to explain what it does and what it can do as accurately as possible, because out of context ā€œsystemd adds sudo cloneā€ people immediately jump to conclusions. It might not be the best idea in the end but itā€™s also worth exploring.

Max_P,
@Max_P@lemmy.max-p.me avatar

Some executables are special. When you run them, they automagically run as root instead! But if sudo isnā€™t very, very careful, you can trick it into letting you run things as root that you shouldnā€™t be able to.

Run0 DMā€™s systemd asking it to go fork a process as root for you, and serves as the middleman between you and the other process.

Max_P,
@Max_P@lemmy.max-p.me avatar

If you dig deeper into systemd, itā€™s not all that far off the Unix philosophy either. Some people seem to think the entirety of systemd runs as PID1, but it really only spawns and tracks processes. Most systemd components are separate processes that focus on their own thing, like journald and log management. Itā€™s kinda nice that they all work very similarly, it makes for a nice clean integrated experience.

Because it all lives in one repo doesnā€™t mean it makes one big fat binary that runs as PID1 and does everything.

Max_P,
@Max_P@lemmy.max-p.me avatar

The same is on the way in the US with how hard conservatives are fighting to keep graduates dumb and educated. Educated people donā€™t lean towards wars.

Max_P,
@Max_P@lemmy.max-p.me avatar

Yeah, even Asahi has better OpenGL support than real macOS. They make damn sure you have to use Metal to get the most out of it, just like eventually you get caught up in DirectX on Windows whether you want it or not. You can use Vulkan and OpenGL, but the OS really wants to work with Metal/DirectX buffers in the end.

I appreciate that the devs care enough to make it really good from the start, because that sets the benchmark. Now the Linux version has to have a similar enough polish to it.

In comparison, Atom and VSCode both worked fine on Linux just about day one thanks to Electron, but it was also widely disliked for the poor performance. Itā€™s a part of what Zed competes on, performance compared to VSCode.

Max_P,
@Max_P@lemmy.max-p.me avatar

Docker would still go through the kernel for the mount, thatā€™s one of the few things Docker canā€™t do because itā€™s the same kernel as the host.

That said I doubt itā€™s been removed from the kernel, only the Samba server. OP is a client.

Max_P,
@Max_P@lemmy.max-p.me avatar

Definitely very subjective. People keep saying macOS has amazing font rendering but for me it just looks like a blurry mess, especially on non-retina displays. My fonts are set to be as sharp as possible on Linux because when coding and in the terminal I want very sharp fonts so theyā€™re easier to read for me.

Seconding the dependence on the particular font as well. Cantarell, Ubuntu and OpenSans are all fairly blurry regardless, unless seen on HiDPI screens in which case they do look more like macOS. DejaVu Sans can be very sharp in contrast at very low resolutions because itā€™s been made in the 800x600 and 1024x768 days and optimized to look sharp when small.

Max_P,
@Max_P@lemmy.max-p.me avatar

What a dystopian world we live on where my 32 thread CPU with 8 channels of 64GB RAM is ā€œobsoleteā€ for Windows 11, because it lacks a fucking TPM of all things.

Max_P,
@Max_P@lemmy.max-p.me avatar

Iā€™ll never understand the people that fake these kinds of things. Fake watches, fake followers, fake views, fake likes, fake jobs. Why?

Whatā€™s attractive about likes and views anyway? Why would I care that my date has 0 followers or a million followers? If anything it means theyā€™ll constantly be busy streaming.

Max_P,
@Max_P@lemmy.max-p.me avatar

Thatā€™s precisely the problem, motherboard manufacturers hoping your CPU won the lottery and can do the +30%. The solution is to tune for the CPUs thatā€™s maxed out of the box.

Reviewers have been complaining about this for years.

Intel publishes recommendations but lets the manufacturers do what they want because it makes their CPUs look more competitive with AMDs, while getting to blame the motherboards when it inevitably becomes a problem. AMD actually specifies maximums and minimums because they donā€™t want people to see their CPUs as unstable, especially when they are ahead of Intel with stock settings.

A motherboard should never overclock by default. Never. If the user wants overclock, they should enable it themselves. Itā€™s an issue dating back two decades, thereā€™s an Old New Thing blog post from Raymond Chen about it where he goes into how they were getting a lot of nonsense bug reports. Turns out some PC shops were shipping prebuilts running right at the edge of stability so once the chips degraded it started crashing random apps. Blog post

Max_P,
@Max_P@lemmy.max-p.me avatar

Theyā€™re not for you. If you want to pump 500W into your CPU, AMD will gladly let you do so. But you did so with your own understanding of the risks.

Those specifications are for motherboard manufacturer defaults. Defaults should always be sane and safe. 14th gen Intel motherboards are shipping with no power limit and increased turbo boost durations out of the box. Those can work, but requires really good cooling that the motherboard have no way of knowing if their users will 100% of the time have watercooled PCs.

People have been ā€œfixingā€ those Intel CPUs by underclocking them where they should always have been in the first place. That shouldnā€™t be underclocking, that should be the stock reference setting.

Max_P,
@Max_P@lemmy.max-p.me avatar

The whole chanting the ā€œaxe the taxā€ to QuĆ©bec where we donā€™t even have the carbon tax was all I needed. Even if we had it, with all our hydro we wouldnā€™t pay much carbon tax anyway.

Heā€™s still managed to convince a ton of the rural population thatā€™s true and that heā€™ll make gas so much cheaper and their profits skyrocket.

Max_P,
@Max_P@lemmy.max-p.me avatar

Itā€™s still pretty popular for most FOSS communities and the only way to get live support by the community, so yes, very much so.

The big advantage is you donā€™t need to sign up for anything, zero terms and conditions to agree too, zero personal information to give away unlike Discord which now wants phone numbers and such for most servers.

Max_P,
@Max_P@lemmy.max-p.me avatar

Been working great for me for ~1 year on my desktop and closer to 2-3 on my laptop.

The only thing missing for me was Barrier for input sharing, which libei is supposed to fix. I ended up going for a hardware solution as Barrier is jank af anyway.

Only thing not working for me is HDR (should be fixed in Plasma 6.1), not like you could do HDR on Xorg anyway. Also no HDMI 2.1 but thatā€™s because fuck the HDMI Forum.

Performance-wise, just blows away Xorg in every metric, and explicit sync should make that even better.

Max_P,
@Max_P@lemmy.max-p.me avatar

Were you using patched KWin or something? Because experimental HDR support is supposed to be one of the big features for 6.0, so unless it broke in 6.0.3 or something, you shouldnā€™t have had an update to break HDR in the first place because it wasnā€™t supported.

Max_P,
@Max_P@lemmy.max-p.me avatar

Well thatā€™s a weird one then. It got released February 38th and took a couple days for Arch to get it. I had the washed out colors too but I didnā€™t have any HDR before that. Thatā€™s ~6 weeks ago so yeah itā€™s probably 6.0.3, the last that came out about that 2-3 weeks ago. I guess you were one of the lucky ones it worked and then broke! With a bit of luck itā€™ll be fixed for good on 6.1.

Max_P,
@Max_P@lemmy.max-p.me avatar

It works for most people but thereā€™s some issues with some monitors where the color saturation doesnā€™t work well and result in washed out colors compared to SDR.

It will also output RGB into YUV buffers if you have a display that only supports YUV colorspaces, so you end up with a very green and reddish purpleish screen.

Initial HDR support was introduced in 6.0, and 6.1 is supposed to bring some fixes for the washed up colors. I havenā€™t found a bug for the YUV stuff and didnā€™t have time to do a proper bug report.

Max_P,
@Max_P@lemmy.max-p.me avatar

A bigger desk so I can just roll the chair a few inches to switch to the work laptop.

My original plan was a keyboard/mouse only KVM, probably a Teensy or a RPi or something of the sorts. But I got lazy as the extra desk space has just made it a non-issue for me. I also have a Logitech mouse that can switch between devices, so if I was going to really need that setup Iā€™d probably just get the matching keyboard.

Anti-web discrimination by banks and online services - is this even legal?

Banks, email providers, booking sites, e-commerce, basically anything where money is involved, itā€™s always the same experience. If you use the Android or iOS app, you stayed signed in indefinitely. If you use a web browser, you get signed out and asked to re-authenticate constantly - and often you have to do it painfully using...

Max_P,
@Max_P@lemmy.max-p.me avatar

Thatā€™s a safety thing. Phones are usually owned by one person or possibly shared in the family, but the security is such that app data is per-user anyway.

Websites though, people still sign in from all sorts of devices and often wildly insecure ones such as public/work computers, one malware away from hackers having access to your bank account.

Inconvenient for advanced users like us, but it would literally make all of those refund scams so much easier to pull off because they wouldnā€™t even have to trick the victims into logging into their bank: blank the screen, transfer the money, tell them their computer is all fixed, bye.

Max_P,
@Max_P@lemmy.max-p.me avatar

If your bank really spies on you through its app, I would change bank. Neither of my bank apps even run in the background or even request sensitive permissions. I will happily change my mind if you can show any proof that this is happening.

Itā€™s purely security. On Windows and largely on Linux desktop as well, any app can easily look at other appā€™s data, thatā€™s why thereā€™s so many browser credential stealers. Maybe youā€™ll never be a victim of this sort of attack, but if it does happen your bank account is gone.

Android and iOS have complete data isolation between apps. Unless you have root on it, even if you install malware and give it the maximum amount of permissions Android can possibly give, it canā€™t access your auth cookies from the bank app. The bank app canā€™t even access them either until you input a pin or biometric data to get it from the TEE.

Thus itā€™s safe for banks to actually let people stay logged in with reduced identification. Browsers canā€™t do that, not without the web integrity.

Weā€™re an absolutely minuscule minority that cares, and could use a stay logged in feature safely in a browser environment.

Dealing with fraud cases is expensive for the banks, they have good reasons to ensure you can only access your bank account under safe conditions. The average person doesnā€™t even know what a web browser is, they know they click the Google and enter what site they want to go to into Google and search for it. Theyā€™re the people that get scammed on the phone. Theyā€™re the people that have their entire life savings wired overseas.

Just let your password manager fill up the login everytime, itā€™s not hard.

Max_P,
@Max_P@lemmy.max-p.me avatar

on a closed-source software stack

Android is open-source. My phone runs an open-source build of it.

At this point itā€™s barely any worse than a web browser. I know itā€™s sandboxed, it canā€™t access anything I donā€™t want to. All it lacks is isolation with the kernel since web browsers run JavaScript and Android runs native code.

Worst comes to worst you just run the app in Waydroid.

Max_P,
@Max_P@lemmy.max-p.me avatar

Iā€™d start by pulling NetworkManager, wpa_supplicant and iwd logs (journalctl -u networkmanager.service --since -2h and so on). Check dmesg as well.

Max_P,
@Max_P@lemmy.max-p.me avatar

Iā€™d still paste it, redacting SSID and MAC addresses.

Even if itā€™s just info messages, itā€™ll still give us an idea of what it does or doesnā€™t do.

I saw the other comment thread about using a static IP sometimes fixing it. We should see that DHCP attempt in the logs. Maybe it succeeds, maybe itā€™s like ā€œbtw the router handed me option 42 and I donā€™t know how to deal with thatā€, maybe itā€™s using the wrong DHCP client (dhclient vs dhcpd) but clues are clues.

Could also tail the entire system log (journalctl -f) so youā€™ll see live output of every service. Maybe it connects fine but something else immediately kicks in and breaks it.

Max_P,
@Max_P@lemmy.max-p.me avatar

Itā€™s still going but I think a good chunk of the FOSS community avoids it. Distros that still ships it disable the telemetry.

Definitely feels like the desperate attempts to monetize it, and the enshittification that typically arises next.

As far as I know itā€™s still fine to use if your distro disables the telemetry, which is what most people had issues with. Itā€™s still under the same license in the end, which is probably why theyā€™re now pivoting to cloud features: that they can make proprietary. Iā€™m sure cloud-based AI plugins are next.

Max_P,
@Max_P@lemmy.max-p.me avatar

Arch is, not sure about the others. I would imagine Debian also is.

Versions 3.0+ of Audacity are affected. Itā€™s not like itā€™s malware and unclean but they did add telemetry and crash reporting and stuff.

Repairing bad sectors in an external drive

So I have this external 2.5" drive salvaged from an old laptop of mine. I was trying to use it to backup/store data but the transfer to the drive fails repeatedly at the ~290GB mark leading me to believe that maybe there is a bad sector on the drive. I tried to inspect the drive using smartmontools and smartctl but since it is...

Max_P,
@Max_P@lemmy.max-p.me avatar

Thatā€™s not looking good, usually on a bad sector the drive will write it to a spare sector transparently and mark it as bad internally. That means theyā€™ve probably all already been used up.

smartctl should work just fine over USB, unless your USB adapter for the drive is really bad. Make sure youā€™re using sudo as well. Worst comes to worst, try using it in a different computer.

Your next goal would be to get it to do a full self test with smartctl. A low level format might help clear some bad state and it might be okay afterwards with a fresh format accounting for whatever defect it built up over time.

I wouldnā€™t recommend it. It might work for a bit and then just die completely.

Max_P,
@Max_P@lemmy.max-p.me avatar

Backup codes. Youā€™re supposed to print them out and put it in a fire safe or something. Theyā€™re longer and not time based and valid until you rotate them. With those you can lose everything and still access your accounts.

My KeePass database is also synchronized locally on most of my devices, so even if my server is dead Iā€™m not really locked out, I just have annoying merge conflicts to resolve.

Also, Yubikeys. Theyā€™re nice. If whatever blackout destroys your Yubikey, you have much worse problems to worry about than checking your email.

Max_P, (edited )
@Max_P@lemmy.max-p.me avatar

Yeah similar setup except I use NextCloud.

KeepassDX is great, can use it with just about anything too. I used it over sftp for a bit. Itā€™ll happily do Google Drive, OneDrive, DropBox and just about anything that implements the right content providers.

Going through the provider is nice, it gives NextCloud an opportunity to sync it before it hands it over to KeepassXC, and knows when it gets saved too so it can sync it immediately. I donā€™t think Iā€™ve had merge conflicts since, and I still have my offline copy just in case.

The annoying part is when youā€™ve added a password on one side and cleaned up a bunch of passwords on the other side. When they get merged, it doesnā€™t merge what changed it merges the databases together so your cleanup is gone. Itā€™s safe at least, and exceedingly rare.

Max_P,
@Max_P@lemmy.max-p.me avatar

Yeah, itā€™ll forward anything that makes its way. Although it doesnā€™t mean you can just proxy anyway through it. If itā€™s on the public Internet for example, sure it would theoretically forward to 10.0.0.0/8 range, but you need a packet addressed to 10.0.0.0/8 to somehow make it to your box in the first place, which you canā€™t do as each hop makes an independent routing decision.

Neighbours on a cloud VPS are definitely the most likely to be able to exploit this, assuming you have a private IP on a shared network somehow and they let you talk to other VMs directly via their private IP. Making a virtual network just for the customerā€™s VMs is incredibly cheap, and most cloud providers either have you make a virtual network or they just come with a default one thatā€™s still all yours, so this is less and less common unless youā€™re on some super old VPS host that did it the lazy way. But even if youā€™re literally on a friendā€™s Proxmox, itā€™s trivial to set up a dedicated virtual network. Even VirtualBox lets you easily make virtual networks.

Iā€™d still set up the firewall though, even if itā€™s just a -A FORWARD -i eth2 -j DROP to explicitly disallow forwarding from the publicly exposed interface, if you really want to blindly trust and forward to internal VPNs and VMs and containers and whatnot you have going that you want to enable forwarding for. But itā€™s really not that hard to set up basic sanity rules like, donā€™t forward traffic coming from the public interface to anywhere.

Things like firewalld or ufw can help you with firewall stuff.

Max_P,
@Max_P@lemmy.max-p.me avatar

To the surprise of absolutely no one.

If you canā€™t share real news people will just share fake news instead.

If you want people to consume quality news from somewhat reputable sources, you donā€™t make it a pain and disincentivize the main sources of traffic from sending traffic that way.

Reliable news are important and they do need funding, but thatā€™s the absolute worst way to go about it. Of course if you put barriers people will cheap out and get what they can get for free. Thatā€™s why we need government funded news like CBC, BBC and others. Itā€™s an essential public service, treat it like one instead of trying to let private companies do whatā€™s most profitable. Quality reporting isnā€™t profitable, shocking headlines are.

Max_P,
@Max_P@lemmy.max-p.me avatar

For the work part in particular, you may find that a virtual machine will get the job done pretty well. With modern CPUs thereā€™s basically no overhead, so itā€™s often easier to just run the OS you need in a VM.

You can just run it in the background, run your .NET and SQL server and stuff, possibly VSCode remote into it. PowerShell over SSH. If you need to run a GUI application, you can just RDP into it. FreeRDP has a cool feature where it can RDP a single application so it pops up as any other app on the host. Works great when you just need the databaseā€™s GUI or whatever.

With virtio drivers and everything, it will be essentially as fast as if native. GUI will be slow because of software rendering, but the point is to use it as a server and only use the GUI as necessary, and for server stuff RDP isnā€™t too bad given itā€™s basically localhost network.

Thereā€™s also the fun but painful world of VFIO and passing through a real GPU into the VM and feeding back the video to the host with Looking Glass. Itā€™s so good you can game on that (anticheat still wonā€™t work though, they donā€™t like VMs), so if RDP bothers you thatā€™s an option. Although in my experience, Visual Studio is just as slow and buggy natively, so I wouldnā€™t bother and just RDP.

If you add virtiofs to the mix, in theory you can share the code folder and use VSCode on the host and compile on the guest, or even compile on the host and run on the guest. Or send compile commands to Windows over SSH from VSCode. I have my entire Steam library (and Proton containers) on virtiofs and it works perfectly, so I have to imagine a code project should work nicely as well.

Virtual machines are an awesome tool. Thereā€™s also the benefit of keeping all the work stuff in a big isolated container. If you have multiple projects you can make multiple VMs and not worry about one project needing version X and the other version Y of whatever.

Max_P,
@Max_P@lemmy.max-p.me avatar

Itā€™s even nice to subdivide some activities. I have a disposable one for running sketchy stuff, two for gaming (Arch and Windows), a few work ones, a few dev ones.

Theyā€™re just so cheap to run these days, I always have one or two running at all times.

Max_P,
@Max_P@lemmy.max-p.me avatar

And then ALSA to all those barely functional audio daemons to PulseAudio, and then again to PipeWire. That sure one took a few tries to figure out right.

Max_P,
@Max_P@lemmy.max-p.me avatar

Wayland is incomplete and unfinished, not broken and obsolete and hopelessly bad design. PulseAudio was bad design. Wayland is very well designed, just, most things havenā€™t been ported for it yet and some design by committee hell, but even that one is kind of a necessary tradeoff so that Wayland actually lasts a long time.

What people see: lol Firefox canā€™t even restore its windows to the right monitors

What the Wayland devs see: so how can we make it so Firefox will also restore its windows correctly on a possible future VR headset environment where the windows maintain their XYZ and rotation placement correctly so the YouTube window you left above the stove goes back above the stove.

The Wayland migration is painful because they took the occasion to redo everything from scratch without the baggage of what traditional X11 apps could do, so there is less likely a need for a Wayland successor when new display tech arrives and also not a single display server thatā€™s so big its quirks are now features developers relied on for 20 years and essentially part of the standard.

Thereā€™s nothing so far that canā€™t be done in Wayland for technical implementation reasons. Itā€™s all because some of the protocols arenā€™t ready yet, or not implemented yet.

Max_P,
@Max_P@lemmy.max-p.me avatar

This has nothing to do with Wayland, itā€™s just AppImages kinda sucking. Use Flatpak or the one in your distroā€™s repos, not the AppImage. AppImages are the equivalent of portable apps on Windows, like the single exe ones youā€™d put on a flash drive to carry around.

Also the AppImage developer is very against Wayland and refuses to support it, which is why Wayland support is a shitshow on AppImages.

If you pick the Flatpak itā€™ll get updated in the background, have a proper launcher and everything.

Max_P,
@Max_P@lemmy.max-p.me avatar

Itā€™s slowly happening. KDE can now do global Xwayland shortcuts, they also implemented XWaylandVideoBridge and compositor restart crash recovery for apps. Weā€™re getting proper HDR, we have proper per-monitor refresh rates and VRR, I can even hotplug GPUs. Some of that stuff works better in XWayland because we can just run multiple instances with different settings. For the particularly stubborn cases, thereā€™s rootful XWayland. X12 would have to break things too, and I doubt an Xorg rewrite would be all that much further than Wayland is. Canonical had a go at it too with Mir which was much less ambitious.

NVIDIA was right on that one indeed, but Wayland also predates Vulkan and was designed for GLES, pretty much at the tail end of big drivers and the beginning of explicit and low level APIs like Vulkan. They could very well have been right with EGLStream too, but graphics on Linux back then was, erm, bad. But in the end theyā€™re all still better than the kludge that is 3D in Xorg.

Itā€™s getting a lot of momentum and a lot of things are getting fixed lately. It went from unusable to ā€œI canā€™t believe itā€™s not Xorg!ā€ just this year for me. Itā€™s very nice when it works well. Weā€™ll get there.

Max_P,
@Max_P@lemmy.max-p.me avatar

40 years old is also what makes it so hard to replace or even reimplement. The bugs are all decade old features, everything is written specifically for Xorg, all of which needs to be emulated correctly. It sure did serve us well, itā€™s impressive how long weā€™ve managed to make it work with technology well beyond the imagination of the engineers in the 80s.

Thereā€™s this for the protocols: github.com/ā€¦/wayland-x11-compat-protocols

It can be done, itā€™s just nobody wants to do it. Itā€™s not really worth the effort, when you can work on making it work properly in Wayland instead. That way you donā€™t need XWayland in the first place, but also XWayland can then implement it using the same public API everyone else does so it works on every compositor.

  • All
  • Subscribed
  • Moderated
  • Favorites
  • ā€¢
  • fightinggames
  • All magazines