Adrian wrote: ↑Sun Apr 27, 2025 11:25 pm
Anything that happens, happens after some kind of upgrade, that's like saying "program failed after I woke up in the morning" unless you downgrade the program and that version works it's just as likely that something else happened to your computer
I doubt this kind of guilt by association is going to go away anytime soon. This logical fallacy has been around so long its name is Latin:
post hoc ergo propter hoc ("after it, therefore because of it"). Home users rarely keep multiple identical systems around to be able to do A/B testing of new features to help pinpoint details like version 1.23 outputs correct data on some input and version 1.34 outputs incorrect data on the same input.
It isn't inherently wrong to say "I updated my machine and now it's broken"; to your point this is an invariable guarantee of sequence. But the critical thinking skills needed to find out if the update broke the thing aren't taught in schools. Sometimes updates really do negatively impact system performance. Therefore it's easy to make the assumption any update can plausibly bear responsibility for any negative change in behavior. We've already seen "I don't understand 'apt search', therefore my next door
neighbor has hacked me". apt-get and dpkg lack an easy rollback mechanism, so users that upgrade and don't like the new packages don't have much recourse to downgrade to a version they do like unless they want to assemble an extensive private collection of .deb files or proactively maintain a point-in-time backup strategy, like snapshots.
Smart people make backups before an upgrade, but Debian doesn't require it and Debian doesn't wrap its installation workflow in anything that would be easy for a beginner to set up -- and fall back to -- a checkpoint that could undo an update if he or she disliked the result.
