Meeting 1 (2022-05-31)

[foolip]

Based on web-platform-tests/interop#41 I've gone ahead and scheduled a meeting for Tuesday, May 31, at 8am PT / 11 noon ET / 4pm UK / 5pm CET. I have invited:

• @jensimmons, from Apple (ET)
• @rachelandrew, from Google (UK)
• @bramus, from Google (CET)
• @skobes, from Google (ET)
• @ststimac, from Microsoft (PT)
• @AdaRoseCannon, from Samsung (UK)

I've made it 30 minutes, because that's unfortunately what I can personally do, but think it's better to get started than continue to struggle with scheduling.

The meeting and following discussion might be similar to web-platform-tests/interop-2022-editing#1, basically first exploring the space to identify the concrete work needed, then switching to execution to improve specs and tests.

Agenda to be posted here.

[bramus]

Just a heads up that @ststimac no longer works at Microsoft.

[foolip]

Thanks @bramus, I didn't know that. I did notice that @karlcow has a new employer however. @karlcow are you still interested in this effort?

[karlcow]

@foolip I let @jensimmons who has been the most active and is already involved, until I figure out things on my side. Thanks.

[foolip]

@kypflug is there someone I should invite from Microsoft?

[foolip]

I promised to post an agenda, better late than never:

• Hello
• Known issues #1, fielding of known issues from other places based on people's experience
• Milestones for this work. Roughly: define the areas we'll endeavor to fix, give each an owner, define our progress (as a score) in terms of those areas
• Cadence and time for meeting

[jgraham]

@emilio @hiikezoe are you the right people to be involved from Mozilla, or alternatively do you know if there's someone else who ought participate in this viewport investigation work?

[foolip]

Brief notes from the meeting

Attending: @rachelandrew @AdaRoseCannon @bramus @skobes @foolip

Brainstormed list of issues:

• Viewport Measurement (including new Viewport Units) interop#4 (comment)
• Viewport Investigation project  interop#41 (comment)
• Known issues #1 (comment)
• For mobile scrolling, when does the address bar animate away, when do events fire, when do units change their meaning?
  • What triggers the resize?
  • How does the page react to the resize?
• Testing this is hard, mobile browsers aren't tested in WPT.
• Virtual keyboard, Safari treats the keyboard as an "overlay" while Chrome shrinks the viewport. https://w3c.github.io/virtual-keyboard/
• Many issues in the CSSWG, go through all of those issues and make sure they're being addressed
• Scrollbars, whether or not 100vw unit shrinks when scrollbar is shown?
• Agree on what vh means, should it be lvh?
• Relationship between viewport(s) and initial containing block
• Samsung Internet might have some differences to Chrome on Android, because it's a different implementation. Needs investigation. New viewport units not enabled for stable in Chromium yet.

@rachelandrew volunteers to look through open CSSWG issues, make a list.

@bramus volunteers to look through differences between browsers in regular vh unit and Visual Viewport API.

[emilio]

@hiikezoe is probably more familiar than me with some of the tricky compat issues, also @theres-waldo perhaps.

But if they can't make it due to timezone issues or conflicts I'm happy to attend this.

[hiikezoe]

I am afraid, it's midnight in my time zone so that I won't be able to join the meeting :/

* Testing this is hard, mobile browsers aren't tested in WPT.

I don't think testing is so hard. Mozilla has been running WPTs on a gecko-based mobile browser, unfortunately the browser doesn't have dynamic toolbar yet (there is an open bug to implement dynamic toolbar in the browser, and fixing it should be straight forward).

And if WebDriver can invoke native touch events, we could have more interesting test cases.

[bramus]

Issue for the 100vw vs scrollbars: w3c/csswg-drafts#6026