But if a CSS property is invoked that the browser doesn't know about, it simply ignores it and moves on. The same goes for HTML tags and attributes. This is less true for JavaScript features because accessing a field of an object or attempting to call a function that doesn't exist will throw an exception and cease execution. Though on sites whose JS is mostly peripheral to the content, this can still sometimes result in a mostly-functional site. You could also play a game where you stub out enough of the APIs to prevent exceptions being thrown, without actually fully implementing the features. i.e. make an API function callable, and just not do anything.
The idea isn't to write off parts of the web entirely, but to be smart about how things are prioritized and aim for graceful degradation of the overall experience, as a way of dramatically lowering the bar for what it takes to make a browser that could be reasonably used day-to-day.
In that case the browser would fail to render or even load some websites, which would convey the idea that the browser was a poor piece of software.