Hi Mark,
thanks for the PR. I'm wondering if removing these lines cause a regression on older versions or other browsers. What do you think?

Good question! I can test with browserstack if we can get a public demo updated.

I've noticed there are other preventDefault usages around, we should fix the issue for all of them

Any news on this? Looking around I've found this one as proposed solution:

(passiveSupported && (active || el == window.document || el == window.document.body || el == window)) ? 
    el.addEventListener(name, fn, { passive: false, capture: bubble }) 
    : el.addEventListener(name, fn, bubble || false);

I admit I haven't had much time to look at this since the original PR... your snippet looks good, I bet that can be abstracted to a function and re-used