Don’t forget when the update stage actually reads 100%, which makes no logical sense because if the stage was at 100%, you wouldn’t still be telling me we are processing it as the current stage.
Somehow Windows struggles with file i/o and always has been. When copying stuff to floppy disks in Win 3 or 95 the progress bar steadily grew to 100% and since floppies were loud, you could hear that the actual copying only started then and you had to wait longer staring at 100% than the progress bar before.
meanwhile updates on linux telling me exactly what is happening in real time, working completely in the background, and politely informing me that i may wish to reboot to apply all the updates properly
Homebrew people: "This patch changed nothing, except they tried to plug a hole. Damn, took us almost 10 minutes to counteract that this time!"
(OK, there was one system update where they added the ability to run stuff off of the SD card, but beside that, there were a whole bunch of updates where they tried to stay ahead of homebrew/pirates and failed spectacularly.)
The old paradox of Microsoft security updates. The more frequent they are, the more they look like they're staying on top of things. While at the same time showing the world there are a lot of frikkin' security holes in Windows all the time.
Part of my job used to involve explaining patch supersedence to leadership so that they had a clear idea of why a totally different patch needs to be loaded to address a vulnerability reporting a different patch number in the scanner.
Hm. That's very generic and seems to only describe the software behavior from a user perspective. When I'm looking for a change log, I'm thinking of something like this: https://github.com/evcc-io/evcc/releases (came to mind because evcc was the last thing I updated).
This comment and subsequent responses are making me wonder now, if you somehow dug out a 15 year old flash drive with like a Firefox 3 installer on it or something, could you get that up and running and eventually updated to the current version?
It's not like they're the first ones to do it either. Ubuntu did it before them and it was a massive disaster. Miscrosoft couldn't not have noticed it. They've seen what happened, and they went "Yes, that's exactly what we want" anyway.
Microsoft also really likes to install the update on your machine, wait a while, then finally activate whatever feature it is they changed.
Like I think I read somewhere that every machine running 22H2 around the time 23H2 came out was actually running 23, but with most of the new features turned off. Also even before 23H3 came out they were sprinkling those features into 22 so by the time I updated nothing changed.
Tinc gets broken by Windows updates every once in a while. The problem is that the update sometimes renames the network connections and Tinc needs the connection to have a specific name to work.
That's the one I personally ran into several times now.
With the millions of updates they keep pushing, you'd think by now, they would have fixed their drivers going stale out of nowhere and shit just stops working until you remove and reinstall them and reboot, for some weird reason.
Wut? If the driver fails for some reason, it gets restarted in the background and you get a small notification in the tray about that. That's all, no need to reinstall anything.
Ooooof. Just this last week I had to remove the drivers for my headsets completely, reboot and then reinstall. Same with the docking station. Shit happens all the time.