News for hackers. Stuff that matters.
Re:CSS Containment Specification by _bxg1 on Sunday February 16, 2020 @02:21AM (#22340336)
And then there's another layer of complication when you assume that any and all styles can change at any time. A big part of "contain"s use-case is when you can guarantee that your JS will never apply certain styles to certain elements. There's no way the browser could ever verify that.
Re:CSS Containment Specification by frivoal on Sunday February 16, 2020 @02:37AM (#22340409)
> when you can guarantee that your JS will never apply certain styles to certain elements

It goes further than that: using contain means that even if you tried applying these styles that would break optimizations, the browsers will not obey you. css-contain isn't just a hint to the browser that you're not screwing things up, it's a mode switch that prevents you from doing so.

Re:CSS Containment Specification by _bxg1 on Sunday February 16, 2020 @02:42AM (#22340428)
True, but the things it prevents you from doing are virtually never something you would want to prevent via a mechanism like this. You'd want to catch them higher up. I don't think it would really be helpful as a constraints system.