I'd encourage you to read some of the testimonials in the bug tracker - the applications this is breaking are not tech-debt-laden monstrosities, they're very simple examples like repl.it and educational coding websites where the IDE is hosted in an iframe and you can run your code in-browser. I can think of many more derelict features I'd have removed first if this was really about cleaning up code smells.

And regardless of whether you think `window.alert` is a giant pile of tech debt: "we don't break userspace" is a mantra that web browser teams would benefit to heed.

It is a mantra that web browser teams heed. That's why the HTML living standard is the absolute mess it is.

"We don't break userspace" is not an absolute rule, and has plenty of exceptions in practice, including security and things where no legitimate use case has been identified.

Linus Torvald coined the phrase "We don't break userspace" and he has been less than nice to people who seek to undermine compatibility in the name of security: http://lkml.iu.edu/hypermail/linux/kernel/1711.2/01701.html

Interestingly enough, the kernel has broken my (userspace) programs at two different occasions. Both times were knowing and intentional, as in the commit messages acknowledged that they were making breaking changes, and they weren't because of security either.

But these were only two occasions over ~10 years, so it's not too big of a deal.

I understand where the phrase is from. I am saying the rule is not as absolute as you think it is. Linus himself breaks userspace on occasion where it is warranted.

But it works. And after, it wont.