Just realising again how an "assumption" can completely skew a representation.
From the Wikpedia page: "The usage share of operating systems is the percentage of computing devices that run each operating system (OS) at any particular time. Most device types that people interact with access the web, so using web access statistics helps compare the usage share of operating systems across most device types, and also the usage share of operating systems used for the same types."
But I take my own house with the general operating systems in use:
1. My primary desktop: Manjaro Linux
2. Wife's primary desktop: Windows
3. Media PC: Linux Mint
3. Open Media Vault machine: Debian Linux
4. Two Pi's: 2x Linux
5. Car infotainment system: Linux
6. Smart TV: Linux
7. My laptop: Windows
8. CCTV NVR: Linux
9. Router: Linux
All the above computing devices are interacted with daily, but actually only two are regularly used for actual desktop web browsing. So that comes out at 50% Linux and 50% Windows (excluding the iOS and Android phones in the house). Of course if I also add the 10+ containers running on the OMV machine, they all run Linux too.
It just makes me really realise how far out these statistics can be, that get thrown about and publicised everywhere. I've long suspected that many Linux OS's in use are not counted because they don't register anywhere centrally, often the browsing data is blocked, and many exist behind the same public IP address. Madagascar, for example, has 32% unknown devices because "the most common browsers are not often used there".
Primary desktop computer use is one thing, but usage share of operating systems is a lot broader, and right now there is probably no way of coming near to estimating it with any reliability at all.
See
Usage share of operating systems - Wikipedia#
technology #
operatingsystems #
statistics #
marketshare The usage share of operating systems is the percentage of computing devices that run each operating system (OS) at any particular time. All such figures are necessarily estimates because data about operating system share is difficult to obtain. There are few reliable primary sources and no agreed methodologies for its collection. Operating systems...
