[–] 2 points 2 weeks ago

If you search through the source, you will find several places explicitly selecting for freebsd. I don't know Go myself, but the first step should be to enable the FreeBSD path on OpenBSD. Currently you're probably in an undefined path (no platform specific code), so fix that first

  • source
  • [–] 1 point 1 month ago

    The first couple sentences are highly alarming. I’d definitely need to know how that happened before plunking my possibly vulnerable code on another OS and assuming that will solve the issue.

    But the rest, yeah, i agree. It’s just simpler and has so many built-ins out of the box.

  • source
  • [–] 3 points 1 month ago

    I should watch the whole presentation when I can.

    If OpenBSD were to do this, OpenBSD would be the perfect server operating system for self-hosting internet-facing services.

  • source
  • [–] 7 points 1 month ago

    This is such a bright spot in tech right now. I would LOVE to have a resilient file system for OBSD. It would open it up to so many more use cases for me.

  • source
  • [–] 1 point 3 months ago (2 children)

    This is very good news as I love and use both software a lot.

    Reading the original post from rsadowski@, it turns out the motivation behind it all is a pretext to use LLMs. Given the amount of patches submitted, I truly hope this code was thoroughly reviewed and remains (even if machine-generated) secure and at the same quality level as what we're used to with OpenBSD software.

  • source
  • [–] 2 points 3 months ago*

    Unlike X, it’s being actively developed

    Wayland is 17 years old. That's beyond mature software, and you still can't RDP into a Wayland session. If you think that's niche, you didn't have a job during the pandemic.

    So ya know, "actively developed" only seems to hold true for flexible definitions of active...

    it might actually gain those missing features

    Yeah okay, let's do that: let's finish Wayland and make it fully usable. I'm 110% onboard with that.

    And to do that, looping back to my original comment, let's not waste energy and time on yet another brand new stack and let's not reinvent the wheel: you want a working Wayland (well, you have it apparently...), I want a working Wayland, everybody wants a working Wayland, and nobody wants Arcan because nobody is interested in reliving another painful transition again, was my point.

  • source
  • parent
  • context
  • [–] 2 points 3 months ago (1 child)

    There are missing or less well-supported features in the wayland protocol as well as the various compositors. I would argue that most of your complaints, if not all of them, are niche concerns for the average Linux desktop user. I can't really say that I've had much trouble with clipboard buffers in various applications (after it worked at all in wayland compositors). I'm either doing something different or there's not much overlap between the applications we regularly use.

    I'm not going to argue that Wayland is flawless or even feature-complete, because it isn't. But for the average Linux desktop user, a decent wayland compositor is completely fine and works well without any need to mess with configs. Unlike X, it's being actively developed so it might actually gain those missing features unlike all the shit X has been missing for ages. I wouldn't even mind another option in the display manager/window compositor world, but your first comment seems way too harsh, especially when most of your criticism boils down to praising the supposed network transparency of X, which has mostly been a hindrance for at least a decade. No wonder they decided to do away with it for Wayland.

  • source
  • parent
  • context
  • [–] 1 point 3 months ago* (last edited 3 months ago) (2 children)
    • Still no decent RDP server for Wayland - VNC only, and only super-hacky.
    • Impossible to inject input events, making all tools that autotype anything pointless.
    • Copy/paste buffer utterly borked in many applications - including most annoyingly Remmina.
    • Nesting Wayland sessions is an exercise in extreme frustration. I know because I figured it out and it's needlessly complicated.
    • Etc...

    Wayland works if you're using your machine for local applications (and not if you use accessibility tools at that...) But as soon as you throw in remote working, networking and multi-users, you're in a world of pain.

    Look, I hate Xorg as much as the next guy, but at least for all its faults, it lets me work. Wayland is constantly in my way because its authors figured "fuck all that networking stuff, let's throw it all away and concentrate on local performances" and now all the stuff Xorg could do (badly, slowly...), Wayland can't do it at all unless you're willing to waste your life hacking around the limitations.

  • source
  • parent
  • context
  • [–] 1 point 3 months ago (4 children)

    I have no idea what kind of criticism this is even supposed to be. All wayland compositors I've used in the past 5 years have worked great. I don't care if they are "mildly faster" for "local graphics", as long as I don't need to deal with X11s shitty configs and endless screen tearing. You couldn't pay me to go back to that sad state of affairs. HDR support is also nice.

    The Xorg transition was always going to be very slow and gradual, but I'm sure people like you who seem to want to actively keep us in this shitty limbo, aren't helping.

  • source
  • parent
  • context
  • [–] -3 points 3 months ago* (6 children)

    It's bad enough that we now have to deal with both Xorg - which still lingers around like the smell of somebody's dead grandmother's decaying corpse - and that turd Wayland that's still half-backed and less functional than X, but mildly faster sometimes for local graphics, after a staggering 17 years.

    If you think anybody is gonna willingly deal with a third stack nobody wanted or asked for, you're dreaming baby.

  • source
  • view more: next ›