Mastodon: @SuitedUpDev@mastodon.online

  • 0 Posts
  • 19 Comments
Joined 1 year ago
cake
Cake day: June 8th, 2023

help-circle


  • I am currently rocking a Shanling M2X which I bought secondhand for €75,- from the local classified ad websites.

    Officially supports up the 512Gb but according some reviews it can do up to 2Tb microSD cards ( https://headfonics.com/shanling-m2x-review/ ). I currently have a music collection of about the same size. The Shanling M2X can drive all the IEMs I throw at it, has enough file system support to read just about anything and has WiFi and Bluetooth. Although the WiFi can only be used to wirelessly sync files (which can be a pro but also a con, depending on how you look at it).

    However if you don’t feel like going secondhand, Shanling offers a whole bunch of devices meant for offline playback of local files.


  • To be completely fair, without the “receipts” (ie, screenshots or something else point towards proof) my comments shouldn’t be considered anything more than hearsay.

    Personally, this experience is something that’s sticking out like a sore thumb. Most people I interact with online, even people I haven’t interacted with before, I start out with the assumption that they come from a place with good intentions. And then there is a person, that immediately goes against that, especially on a platform where I didn’t expect it.

    Edit: grammar



  • Closed source isn’t a crime. However trying to ruin a company with exclusionary tactics can be. Linux kernel devs and Wayland devs have all conspired to harm a company.

    NVIDIA kinda shoot themselves in the foot on Linux and excluding themselves. Refusing to support generally supported APIs like;

    • VA-API

    NVIDIA rather wants the OSS community the use their VDPAU or NVENC / NVDEC API’s. Whilst everything and dog uses VA-API.

    • GBM

    Not true anymore (for driver above 495), but in the past NVIDIA refused to support GBM (for Wayland) and rather have compositors use EGLStreams instead of GBM.

    Next to that modern NVIDIA hardware (GTX 900 and 1000 series) on the opensource Nouveau drivers cannot be reclocked because it needs some magically blessed signature by NVIDIA. NVIDIA refuses to supply that signature for that hardware but did release it for 1600 and up series.

    That’s just two things where I am like, dafuq are you doing NVIDIA…


  • SuitedUpDev@feddit.nltoLinux@lemmy.mlHyprland is now fully independent!
    link
    fedilink
    English
    arrow-up
    46
    arrow-down
    3
    ·
    4 months ago

    I whole-heartedly agree with this one and I am genuinely not surprised about the behaviour of Vaxry.

    To give some context around this, ThatOneCalculator (aka Kainoa, the person behind Firefish) and I maintained the AUR package for hyprland-git back in 2022. When I initially made the AUR package file, it wasn’t great (and there were a lot of points to improve these packages) but it worked mostly. Of course there were edge cases where building broke, especially this was my first bigger AUR package to maintain. With it being a -git package in the AUR, breakage is to be expected.

    Fast forward about a month, a month and a half. Hyprland rolled out some big changes which caused some build errors. But because my personal life got in the way, Kainoa got sick (IIRC) and I had troubles getting the build scripts working again, so it took a few days to get this resolved.

    Vaxry came complaining to comment section of the AUR package “when are you gonna get of your lazy ass and fix this shit” (or something similar to that meaning, I can’t find the original comment anymore). After that, I promptly disowned the package and let Vaxry handle it himself.

    Because fuck that shit, as package maintainer, I refused to be treated like this. If you think it takes too long, sure, fine, ask if I need help, offer support, anything. But just don’t be an asshole towards people, that offer your software to a wider audience.


  • Yes you can, it won’t be great though.

    I used to maintain a Linux distribution called “OpenWM8650” (back in 2011 / 2012) which was specially aimed at the WM8650 and WM8505. It would run off the SD card. Which wasn’t great, but the flash onboard support was horrible at best.

    Maybe you can find some old information on it, on XDA because the website for the initial distribution is long gone.


  • Debian Woody > Red Hat 7.0 > Slackware 9.0 > Slackware 10 > Debian > Ubuntu > Mac OS > Ubuntu > Arch.

    At least for desktops and laptops.

    For servers I’m still primarily running Debian (and one instance I’m running Arch).

    The reason why I settled on Arch is primarily because the combination of bleeding edge and being stable enough for daily driving it. The AUR also adds sooo much, that there is nothing I really don’t need to manually install anymore.

    For servers, I basically want a rock stable system. Hence why I’ve chosen Debian Stable.


  • For all the oldies in the room, Lindows / Linspire (me included)

    Trying to merge Wine with Linux into one ““convenient”” package was overselling it at best and worst a lie. Back in '08 / '09 I tried to run it, just to see how good (or bad) it was. It was basically Debian with KDE and their special sauce on top to ““easily”” run Windows binaries. Especially in those day, WINE wasn’t as good as it is today. I think you can fill in the rest ;-)





  • Oh definitely that the next couple of weeks will be spent preparing for a new wave. I wouldn’t be surprised if that was the reason why Lemmy.ml was down. Yes I know that it was because of a server migration but, that server migration also didn’t happen for no reason ;-)

    The first migration wave to Mastodon was very bad. I still remember, that following a person was nearly impossible. Not to mention the fact that, even if you could follow someone, it didn’t automatically meant their messages ended up in your feed.

    I know for example that version 0.17.4 of Lemmy (that was released 2 days ago) already has some database optimizations.



  • In theory, I agree with you! A 100%, but the problem is that currently Lemmy doesn’t support migrating your profile to a different server. So that already slightly complicates things. So from the get-go they are forced to make choice. A choice which isn’t clear, what potential consequences are and the fact they currently easily migrate to a different server, obviously doesn’t help.

    “Like email” is basically the same description I’ve been using to explain it to non-tech people.

    Long story short, onboarding needs to get better. But that also applies for other Fediverse projects (like Mastodon or Friendica).



  • In short, yes it will take longer to “catch up”.

    The longer answer, be patient. This is an issue that will be resolved. I’m sure that the admins are doing their best to get the instance up & running ASAP.

    From my experience during the great Twitter migration to Mastodon, you saw the same thing. Especially the “main instance(s)”, mastodon.online and mastodon.social in this case, were overloaded to the brim. After a day or two, the sysadmins and developers were able to stabilize the instance.

    The sysadmins in this case were responsible to keep the servers up and running (and scale up where needed). The developers because their code hasn’t been “battle tested” for this level of interaction. So changes were needed, to optimize and squash bugs.

    I hope this helps 😊

    Edit: fat fingers, ie fixed typo