Not only that but looks to be via tracking code used on various sites. I suspect a larger percentage of Linux users will be using ad blockers that likely block that further reducing the numbers.
The stack overflow survey puts the various Linux distros at about 58.4%* with windows at 59.72% and MacOS at 32.57% for personal use. This both accounts for people using multiple OSs and gives a much more accurate number - at least among developers. Though I do suspect these numbers will be higher for developers than the general population but 3% to 58% is a huge discrepancy.
The steam hardware survey does put the number at 1.44%, though that is extremely biased towards windows for obvious reasons. And I don’t know of any more accurate ways of counting then these two. But they only show two wieldy different domains.
* wonder why they broke that out to different distros this year pushing it down the list by splitting it up 🤔 why not also report on the different versions of windows? 😒
I don’t think you can just add the linux percentages, lots of people are probably using several distros. But the 27% of ubuntu still give a much higher lower bound, and while developers are not a very representative subgroup, I’d also estimate the total use percentage a little higher than 3%.
I think it’s browser user agents.
Which have been wildly unuseful for over a decade
Not only that but looks to be via tracking code used on various sites. I suspect a larger percentage of Linux users will be using ad blockers that likely block that further reducing the numbers.
The stack overflow survey puts the various Linux distros at about 58.4%* with windows at 59.72% and MacOS at 32.57% for personal use. This both accounts for people using multiple OSs and gives a much more accurate number - at least among developers. Though I do suspect these numbers will be higher for developers than the general population but 3% to 58% is a huge discrepancy.
The steam hardware survey does put the number at 1.44%, though that is extremely biased towards windows for obvious reasons. And I don’t know of any more accurate ways of counting then these two. But they only show two wieldy different domains.
* wonder why they broke that out to different distros this year pushing it down the list by splitting it up 🤔 why not also report on the different versions of windows? 😒
I don’t think you can just add the linux percentages, lots of people are probably using several distros. But the 27% of ubuntu still give a much higher lower bound, and while developers are not a very representative subgroup, I’d also estimate the total use percentage a little higher than 3%.