Talk:Usage share of operating systems

101% market share for desktop OSes

The market share percentages for desktop operating systems add up to more than 100%. 82.78.167.3 (talk) 00:29, 8 June 2025 (UTC)

And now they add up to less than 100%. As per what I added to the caption, "Percentages may not add up to 100% due to rounding."; that's rounding up or down. Guy Harris (talk) 23:21, 5 July 2025 (UTC)

Web clients

The W3Counter data should split from the other sources in different table, since it uses very different data structure compared to StatCounter or Wikimedia. Provided with a note that W3Counter shows only the first 10 most registered OSs for the period - hence the hefty 'Others' percentage left.

While StatCounter and Wikimedia could stay together, because they both follow very similar data structure. And only then would be easy the data from StatCounter and Wikimedia (both sources published newer or up-to-date data that should be updated inn the article, but also, at the moment is rather messed with the W3Counter data). --Svilenov (talk) 11:40, 23 November 2025 (UTC)

Needs clearer explanation of methods and sources: IoT, and various "embedded," and "headless"

IoT equipment are almost universally embedded Linux, or Android. A similar statement can be made about home multimedia set top boxes like Tivos (or anything even remotely resembling a Tivo since ~'05), physical media players (eg Bluray), and various "smart" equipment (Smart receivers); Virtually 100% of this market uses embedded Linux. TVs, which have by '26 almost universally become "Smart TVs" with few exceptions, almost universally run Android. There is some question as to where to draw the line between "android based" and other embedded Linux OS variants. For example: Fire OS (Amazon stuff). By most definitions, this is just Android. However, its unclear if the sources of the reprinted or used data understand the subtlety here (It seems like they don't, and take Trademarks and Revenue at face value rather than considering the technical distinction)~2026-10043-38 (talk) 08:12, 16 February 2026 (UTC)

Reliability of StatCounter

Is there anyone that can defend that reliability of StatCounter? It's one thing that several ytimes in the past year its had some unknown party spike the reported usage of Windows 7 - at first I chalked this up to just a mishap that they usually fix, until they took a full month (or was it two?) to fix it. Now, steadily climbing the ranks of OS competition is "Unknown"; this was always there but it was never previously at nearly 20%.

This site was never 100% reliable due to how its collected its data, but it seems its flaws have only become more apparent as of recent. I suspect somebody, intentionally or unintentionally, is actively messing with the stats, and StatCounter has shown they are not proactive in fixing it.

Not an editor here but I do want to see people's thoughts. IoI xD (talk) 00:49, 21 April 2026 (UTC)