Why is the expectation that security and feature updates can not be completely separate?
That's the question, now here's the bit you need to read before the answer. I imagine the problem is because things aren't built separately; updated code in one place expects to communicate with an updated version elsewhere. I didn't take compsci classes, there's probably already a documented explanation to this, specifically how feature and security changes can't be isolated. It probably has a name too.
But in my head I'm thinking, this is a problem that could be designed around. It's just that the extra work is substantial and requires a level of organization and... discipline? That no company that pays it's programmers would be willing to put up.
Back on the topic of "someone probably already researched this", I'm sure there's a word for the constant feature and UI churn that most programs succumb to nowadays as well. I know we all like "enshittification" around here but I'm sure it's been going on for FAR longer than that.
All of that being said, serious answers only please. "Companies dumb and don't do shit right" may be correct but it's not helpful if I'm actually trying to pin down the words for all of this stuff.