I've known for a while but Linux really isn't newbie friendly. I've tried several times over the years. Most recently I tried Mint and within a week or two the GUI software update stopped working. I switched to Ubuntu because I have a little familiarity and quickly realized the only way to stay updated is apt. Firefox will work for a while and then have days where almost every tab is crashing constantly. Sometimes you can get a page to work with reload or two, other times not. So I installed chrome under protest as a backup. Mostly it doesn't crash although sometimes it does the same thing as Firefox (but usually not at the same time). Those are really my only complaints at all.
I will say, though, that gaming has come a long way, especially with Steam. I don't think SteamVR is there yet, but otherwise I haven't found anything that wouldn't run (but I play AAA on PS so I'm not really pushing it hard). I thought I was giving up gaming when I moved to Linux, but not at all.
I left a big partition free in the expectation of dual-booting windows, but I really haven't had any reason to at all.
I have been attempting to use Linux for 20 years now. It has found its niche used for me over the years. For example when my kids were toddlers they had a old machine that defaulted to PBSkids. Before that I used it to run a gaming server.
Currently I have a old laptop that I dual boot with win 10 and whatever Linux distro I feel like trying at the moment.
The win 10 on the laptop barely meets the minimum hardware requirements and takes 10 minutes to load.
I have tried a few different distros and always had a few issues with the setup. All sorts of different ones - screen orientation, WiFi connection, printer hell, keyboard layout etc. Takes me days to fix the bugs or give up.
Mint takes 2 minutes to load and so far is working seamlessly. It's apt manager is the easiest I have used in a Linux distro. It found my network printer automatically. It runs smoother than windows 7 did on my laptop.
With Microsoft ending win 10 for the shitty win 11, I imagine many people are looking for alternatives. If Mint continues to work to make setup and usage easy, it will gain market share rapidly. It's not all the way there yet, but it's a hell of a lot better than before.
Try VanillaOS or Fedora Silverblue. They are immutable distros so breaking them are harder than usual. (OpenSUSE was also making one iirc) If you just want something that works out of the box, try them.
Just a word of warning, Windows may fuck your dual boot plans up with an update. I lost all the data for my filesystems on my main drive with a Windows update. It decided to delete grub and the partition data. I recovered a lot of it, but I was never able to get Windows to boot again. That's why I'm 100% Linux now. I just decided Windows wasn't required anymore and installed everything fresh, and gave 100% of the drive to Linux.
I would assume Steam to be lower because gaming was one of the major roadblocks for gaming until recently, and many gamers are still scared of switching over. It also assumes no dual booting and whatever they ran the check on is the registered OS. This would register both potentially.
Team's data is also useful, but it's different data.
I dual boot. I use Linux most of the time for everything, but I switch to Windows whenever I want to play on Steam. I just don't have the time to bother with abstractions layers, drivers and whatnot, even if I read that Steam makes it easy to run Windows games on Linux now.
Windows is more prevalent with gamers so I guess it would make sense for the Steam survey to show less Linux users than there are in the general population.
What do I make up? I guess what the uncertainty could be and saying that they shouldn't give such a precise number, if it's in reality not that precise.
Ehh some linux setups may go into unknown but linux users are also more likely to own multiple devices. I dont think its that precise but its impossible to make it precise
We measure internet usage trends. To accurately measure usage, we have to base our stats on page views (and not unique visitors). Let's look at an example:
Person X uses two browsers. On a particular day, they load one page in Browser A. They load 500 pages in Browser B.
If we based our stats on unique visitors then usage of Browser A and B would both be recorded as 50%. This is obviously incorrect. This does not fairly represent the usage of the browsers given that Browser B was used 500 times more than browser A.
Using page views as the basis of our stats means that Browser A will be recorded at less than 1% in our stats whereas Browser B will be recorded at over 99%. In our view, this gives a more accurate representation of actual browser usage.
I love the fact that when some companies make software they just casually ignore 25% of the world. Why the hell is there so much software that only supports windows. I know its because people are lazy and they cant code but its still really annoying