https://gs.statcounter.com/os-market-share/desktop/north-america/#monthly-200901-202607
Cloudflare stats also seem to support peaks of 15% usage but around 9-10% on average. In the United States alone, the market share for Linux is nearly 12% now for the whole month and Cloudflare stats suggest peaks of up to 25% (!) Linux desktop usage on certain days and times.
The trendline suggests this is just the start. When will companies realize they cannot ignore Linux anymore?
So I can’t find a great answer, but as best I can tell StatCounter uses user agents to determine this data, the web browser being the most obvious source but not the only source.
The user agent for most web browsers on “macOS” the operating system generally report as “Mac OS X”. So other user agents, which report things like “Macintosh” but rarely actually “macOS”, are reported as “macOS”.
This also explains why “OS X” has a larger share than “macOS”. Web browsers on the macOS operating system are used more than “other”.
At first glance you’d think “OS X” and “macOS” should just be combined. But chances are if someone is using a web browser on macOS, they’re probably also using “other”.
Another problem is that we don’t know how StatCounter is handling duplicate/fake data. If I constantly hit refresh or send analytics from a single device, how is that counted? What if I send “OS X” and “macOS” data?
Luckily we also have the Cloudflare data, which shows Linux between 9% and 15%. So StatCounter using 10% is probably on the lower side, but within the margins.