Upcoming HTML standard issue triage meeting on 9/2/2021

[past]

Today we held our monthly triage call (#6884) and I will post the meeting notes there in a bit. The next one is scheduled for September 2nd, 9 am PST. People interested in attending the next call please respond here or reach out privately to me or the spec editors. We will be tagging issues for the next call again using the agenda+ label and we would like to invite anyone that can contribute to said issues to join us.

[edent]

I'd like to attend and discuss #6952 if that's OK?

[past]

I have added #6952 to the agenda, but I cannot promise that we will get to it this time. The list of discussion topics is large enough and we already have 3 carryover items from last time. Would you like to join regardless or are you interested in providing more context in the (unlikely in my view) case of disagreement?

Have you considered just fixing your PR with the proposed changes instead and getting it landed?

[past]

Hey @whatwg/html-triage, our next meeting is coming up in 2 weeks. Please add items to the agenda by using the agenda+ label on open issues or by filing new ones.

[past]

Thanks everyone for attending! Here are the notes from this meeting (next one is at #7014):

Agenda

1. [Panos] Process question: would it be disruptive if we removed existing agenda+ labels after each meeting (except for any carryover items)? This way the slate will be clean for the next one.
   1. Agreement to clean the slate after the meeting.
2. Review past action items
   1. Domenic will take point on moving forward in the remaining bits for removing plugins, there is appetite to continue here.
      1. Progress was made
      2. <param> is most in need of feedback, discussed it later in the meeting
   2. Everyone is encouraged to provide input on this issue in Open UI.
      1. Several comments have been posted.
   3. Agreement that we need WPT tests. Joey will file an issue for WPT support for find in page (which sometimes behaves differently from window.find).
      1. Done!
   4. Domenic will kick off a survey of all pickers and how they can be opened, how click() behaves, etc.
      1. Results
      2. Domenic will do extra testing for browser behavior with no user activation. Consensus that the proposal is reasonable. Also need to think about whether third-party frames can show pickers. (Modals are on top of other pages, file pickers are particularly scary.)
   5. [Directionality in Shadow DOM] Eric will provide a normative definition for directionality inheritance. Brian will seek feedback from CSS WG folks. Eric will respond to Olli's question in the GH issue.
      1. Eric responded to Olli.
      2. CSS WG feedback is still pending, now targeting next week's call. Eric will present a proposed text change to the WG. Panos will keep the agenda item.
   6. Mason will document what Chrome does in back/forward cases vs. document/tab close/restore cases. Melanie will bring this topic up with OpenUI folks.
      1. Mason posted some findings.
      2. Mason will roll back the recent Chromium change.
      3. Domenic to send a PR to clarify the spec about not firing events in these cases.
3. Carryovers from last time
   1. [Domenic] Add &nnbsp; entity for U+202F
      1. Consensus that the downsides outweigh the benefits here.
   2. [Domenic] setCustomValidity() seems pointless on
      1. We'd like to see a little more developer support for this. There is also some small compat risk that is being considered.
   3. [Domenic] Consider creating an HTML search element
      1. The group is generally supportive, Domenic will follow up in the issue. There is a small concern about potential confusion if using the role name "search".
4. [Domenic] Improve autofocus and delegatesFocus interaction
   1. People are encouraged to review and comment on the PR.
5. [Anne] Ensure PDF browsing contexts have no DOM
   1. Mason will look into it.
6. [Domenic] Newly created browsing contexts will almost always reject history.pushState/replaceState
   1. Domenic will take Anne's counterproposal to Chromium for consideration.
7. [Domenic] Define focus behavior for <dialog> element with Shadow DOM
   1. There was agreement to proceed according to the latest findings in the issue.
8. [Domenic] <object> without "data" attribute probably needs to consider params to figure out its URL
   1. Mason will add a use counter.

Action Items

1. Panos will clean the agenda labels after the meeting.
2. Domenic will do extra testing for browser behavior with no user activation. Consensus that the proposal is reasonable. Also need to think about whether third-party frames can show pickers. (Modals are on top of other pages, file pickers are particularly scary.)
3. On Directionality in Shadow DOM, Eric will present a proposed text change to the CSS WG and get their feedback.
4. Mason will roll back the recent Chromium change around element restoration and Domenic will send a PR to clarify the spec about not firing events in these cases.
5. Domenic and Anne will respond to the &nnbsp; issue.
6. On custom validity checks for fieldsets, the group would like to see a little more developer support. There is also some small compat risk that is being considered. Domenic will try to get more signals.
7. The group is generally supportive of a search element, Domenic will follow up in the issue. There is a small concern about potential confusion if the name "search" is used.
8. People are encouraged to review and comment on the Improve autofocus and delegatesFocus interaction PR.
9. Mason will look into the issue about ensuring PDF browsing contexts have no DOM.
10. Domenic will take Anne's counterproposal on Newly created browsing contexts will almost always reject history.pushState/replaceState to Chromium for consideration.
11. Mason will add a use counter for <object> without "data" attribute probably needs to consider params to figure out its URL.