No. They should burn it.

Yes, unironically. It’d eliminate most security risks currently inherent to Windows while allowing for decent customization. It would also greatly accelerate WINE and ReactOS, giving legacy software a smooth offramp from insecure older versions of windows to Linux.

Yes.

I’d expect the author of this article to understand what the word embraced means to start with. If this is m$ embracing OS, it’s akin to seeing two people shake hands and shouting “OMG they’re fucking!”

If you go deep enough, there’s still windows 3.11 dialog boxes in Windows 11 for some core functionality.

https://web.archive.org/web/20240102024352/https://ntdev.blog/2023/01/01/state-of-the-windows-how-many-layers-of-ui-inconsistencies-are-in-windows-11/

An interesting assertion. A full install of 3.11 was about 8 MB or so, and all of the 8086 / -186 / -286 / -386 code will have been thrown away a long time ago. I doubt there’s much of PROGMAN left, and all the fonts and art assets are long superseded. So in terms of total code, it can’t be much. But on the other hand, the code that you write for an event loop or to handle driver interrupts hasn’t changed conceptually very much in that time. Most programmers would reimplement the basics in a very similar way, so there’s not much point in redoing it.

When I used to work in the water industry, we still had programmable logic controllers (PLCs) controlling pumpsets from the 1950s. The last person that could have modified them had retired and since died more than 30 years before. But deciding which pumps to run in order to best fill a reservoir is not logic that needs updating every day, not even every decade. Still working fine, don’t touch it. So I still laugh at my colleagues that can’t touch code that was written a few years ago in an unfashionable library. That’s not tech debt. Try, written by your grandparents for CPUs that had stopped being made before you were born.

And I remember 3.11 being perfectly good enough at the time, anyway. Wasn’t any Linux at that point.

In new displays as well? Or are there some legacy considerations for the old displays that are still around?

I saw a Windows OOM error screen (something modern, win 10 at least) on my local transit providers in tram screen. One of them, the other 10 were working fine. That means that they each screen has its own os and hardware that needs to go with it.

How is that financially dependable or even practical to maintain? Hardware solution: video signal splitter Software solution: embedded controller board (pi1 would be good enough or some industry comparable product) for own os and software to run the information grabbed from the network. As images or as data that is then rendered.

I get that supporting legacy sucks. But it must be cheaper (not to mention easier and good on the brains for the it people) to get this stuff updated and kick it out. I kinda get it with banking mainframes and such but in this case there is nearly no risk involved when it goes down and needs to be fixed. And it went down anyways.

Most german thing I’ve heard today