It peaked at 4.05% in March. The last 2 months it went just below 4% as the Unknown category increased. For June the reverse happened, so 4.04% seems to be the real current share of Linux on Desktop as desktop clients were read properly/werent spoofed.

  • Fonzie!@ttrpg.network
    link
    fedilink
    arrow-up
    84
    arrow-down
    2
    ·
    3 months ago

    In all seriousness, I think government bodies switching to Linux (UK’s, China’s, some Indian states’) attributes the most to this.

      • leopold@lemmy.kde.social
        link
        fedilink
        English
        arrow-up
        19
        ·
        3 months ago

        Source? Last I checked, the Steam Deck was very much in the minority even when narrowed down to just desktop Linux.

      • 737@lemmy.blahaj.zone
        link
        fedilink
        arrow-up
        15
        ·
        3 months ago

        no, the statistics are based on browser agents, very few steam deck users browse the Internet on their devices. it’s also only half the Linux devices on steam, not of all Linux desktops

      • wischi@programming.dev
        link
        fedilink
        arrow-up
        13
        arrow-down
        3
        ·
        3 months ago

        But that’s not really a Desktop is it? If we’d count mobile device we’d also have to include Android and then the situation would look completely different.

        • Bulletdust@lemmy.ml
          link
          fedilink
          arrow-up
          8
          arrow-down
          1
          ·
          3 months ago

          Connect the Steam Deck to a compatible dock and you can quite easily use it as a desktop. At the end of the day, it’s still an x64 based PC that’s just handheld.

          • wischi@programming.dev
            link
            fedilink
            arrow-up
            2
            arrow-down
            2
            ·
            3 months ago

            I’m not sure that’s really a good argument. I can connect an android smartphone to a monitor, keyboard and mouse and call it Desktop. It’s also just an arm64 or x64 based PC just handheld.

            A Desktop PC IMHO is a device that is used for everyday “office” work and neither android smartphones nor steamdecks are that - but laptops for example are (IMHO)

        • freebee@sh.itjust.works
          link
          fedilink
          arrow-up
          2
          ·
          3 months ago

          It is. I use it as such regularly. Keyboard+mouse+screen = browsing firefox as usual. Works quite well. Libreoffice, okular, signal desktop… I’ve used worse computers in recent years, steamdeck desktop experience is better than many 4 to 5 year old cheap laptops with win10 or win11.

      • istanbullu@lemmy.ml
        link
        fedilink
        arrow-up
        8
        ·
        3 months ago

        In Steam maybe. But this is StatCounter which is website visits. I doubt many Deck users are browsing the web.

        • GolfNovemberUniform@lemmy.ml
          link
          fedilink
          arrow-up
          1
          arrow-down
          1
          ·
          3 months ago

          For some reason I think a lot of them (probably even more than half) have tried browsing the web or at least using the desktop mode at least once.

    • Vincent@feddit.nl
      link
      fedilink
      arrow-up
      17
      ·
      3 months ago

      I’m fairly sure it’s deficiencies in StatCounter’s measurement that’s accounting for it. Statistical noise, basically.

      • doodledup@lemmy.world
        link
        fedilink
        arrow-up
        22
        ·
        3 months ago

        It’s probably even higher than that. These stats are mostly based on website visits I believe. And many Linux users are also privacy-minded and might spoof their OS in the browser. I bet a large portion of the Unknown is actually Linux too.

        • Vincent@feddit.nl
          link
          fedilink
          arrow-up
          10
          ·
          3 months ago

          It’s hard to tell, as there are so many things that influence it. A huge factor is selection bias, as only a small number of website embed StatCounter, and that’s very likely to not be a representative sample. I’d bet that the influence of that is magnitudes larger than of user agent spoofing.