alt Hacker NewsTbh i'm starting to think that I do not see microsoft being able to keep it's position in the OS market ; with steam doing all the hard work and having a great market to play with ; the vast distributions to choose from, and most importantly how easy it has become to create an operating system from scratch - they not only lost all possible appeal, they seem stuck on really weird fetichism with their taskbar and just didn't provide me any kind of reason to be excited about windows.
Their research department rocks however so it's not a full bash on Microsoft at all - i just feel like they are focusing on other way more interesting stuff
Replies
First Valve has to actually start pushing for proper Linux games, until then Windows can keep enjoying its 70% market share, with game studios using Windows business as usual.
Also Raspeberri PIs are the only GNU/Linux devices most people can find at retail stores.
> Tbh i'm starting to think that I do not see microsoft being able to keep it's position in the OS market
It's a big space. Traditionally, Microsoft has held both the multimedia, gaming and lots of professional segments, but with Valve doing a large push into the two first and Microsoft not even giving it a half-hearted try, it might just be that corporate computers continue using Microsoft, people's home media equipment is all Valve and hipsters (and others...) keep on using Apple.
Add to that all the bullshit they have been pushing on their customers lately: * OS level adds
* invasive AI integration
* dropping support for 40% of their installed base (Windows 10)
* forcing useless DRM/trusted computing hardware - TPM - as a requirement to install the new and objectively worse Windows version version, with even more spying and worse performance (Windows 11)
With that I think their prospects are bleak & I have no idea who would install anything else than Steam OS or Bazzite in the future with this kind of Microsoft behavior.
Kernel improvements are interesting to geeks and data centers, but open source is fundamentally incompatible with great user experience.
Great UX requires a lot of work that is hard but not algorithmically challenging. It requires consistency and getting many stakeholders to buy in. It requires spending lots of time on things that will never be used by more than 10-20% of people.
Windows got a proper graphics compositor (DWM) in 2006 and made it mandatory in 2012. macOS had one even earlier. Linux fought against Compiz and while Wayland feels inevitable vocal forces still complain about/argue against it. Linux has a dozen incompatible UI toolkits.
Screen readers on Linux are a mess. High contrast is a mess. Setting font size in a way that most programs respect is a mess. Consistent keyboard shortcuts are a mess.
I could go on, but these are problems that open source is not set up to solve. These are problems that are hard, annoying, not particularly fun. People generally only solve them when they are paid to, and often only when governments or large customers pass laws requiring the work to be done and threaten to not buy your product if you don't do it. But they are crucially important things to building a great, widely adopted experience.