> The only time I’ve had a kernel update break things > That definitely broke my assumption that kernel updates were well vetted for regressions. Shouldn't th

> The only time I’ve had a kernel update break things > That definitely broke my assumption that kernel updates were well vetted for regressions. Shouldn't that be evidence for the contrary? One break in a decade or two of use isn't so bad. It's actually pretty good.