David Bokan

Awesome, congrats! I'll try to give it a look next week if I can find some time. Any reason why the visual viewport API is limited to Android? IIRC desktop...

That's great news, thanks for your work here! I'll try to play around and see if I can spot any issues and update the documentation. BTW, is Firefox implementing the...

Your observation is correct w.r.t. Chrome's behavior Whether that's the right thing, I'm not sure. The ICB size is shrunk by the window's scrollbar so in that sense it's not...

Yes, I this is a bug in Chrome: crbug.com/914450. IMO, ScrollIntoView should also scroll the visual viewport but doesn't in Chrome today.

Yeah, there's definitely cases where it shouldn't activate. However, I'm a bit confused why the URL bar was hideable at all; normally the page has to have scrollable overflow or...

Thanks, that's a good case to keep in mind. I chose to keep it restricted at first since it's much easier to loosen restrictions later than it is to make...

So your element is normally responsible for scrolling (in Chrome). Unlike most scrolling boxes, it doesn't establish its own clip (it's actually overflow:visible IIRC), but rather gets clipped by the...

No, not today (unless there's a bug, I actually haven't tried it with a border yet). You could define what to do in that case but I think in general...

Amp also uses a header. The way they make it work is (as Jake mentioned) by making the header position: fixed and applying content padding to the scroller. (e.g. http://bokand.github.io/totese.html)

Yeah, I'll try to compile some more thorough arguments - though allowing the URL bar to show/hide isn't a browser bug but innate in how the web works today. That...