W3C

– DRAFT –
PEWG

20 November 2024

Attendees

Present
mustaq, Patrick_H_Lauke, plh, smaug
Regrets
-
Chair
Patrick H. Lauke
Scribe
Patrick_H_Lauke, Patrick H. Lauke

Meeting minutes

Limit the precision of floating point event fields w3c/pointerevents#517

Patrick: rob wanted to work on this, but it IS slightly non-trivial. see there's activity just today on it w3c/pointerevents#517 (comment)

Patrick: looking at the web animations wording, yes i think using wording of that nature, mentioning potential privacy issues, and then reiterating that again in the privacy section, might be the way to go

Patrick: we don't want to unnecessarily limit the precision IF a web application actually wants to take full advantage of precise sensors. looking at angles, even capping to 1 degree seems excessive

Patrick: don't think Rob has had a chance, so I might take it off his hands and make a first stab at it. i envisage a separate section after all the values in the IDL (rather than mentioning it for each value). for position related things, saying it should at least resolve to hardware pixel. for others, might be a bit trickier. and explain the core concern here about fluctuations in sensors that are consistent and can be us[CUT]

ingerprint

ACTION: Patrick to draft PR for #517

Patrick: On that topic, it seems we're still waiting for wide review from security w3c/security-request#67

Patrick: Philippe, what's the process on this? can we leave it after various attempts to get it?

PLH: yes, you can time out

PLH: how long ago did you ask?

Patrick: May

PLH: then you can time out without chasing them up. if it comes to it later, we may ask them again, but for now, we can time it out

[touch actions] handwriting manipulation type to distinguish panning w3c/pointerevents#516

Patrick: Olga had some actions from the last meeting, no action from us at the moment

Olli: looks like there's a PR for their explainer, but it's still open

<smaug> MicrosoftEdge/MSEdgeExplainers#903

Meta-issue: update WPT to cover Pointer Events Level 3 #445 w3c/pointerevents#445

Patrick: I think we're in a good place for this one https://github.com/w3c/pointerevents/issues?q=label%3Aneeds-wpt+

Mustaq: Rob has been working on #300 - have some background about it: test failed due to some fishy behaviour on Chromium side, should be an easy fix though

Patrick: then for

Patrick: #509 i looked earlier and added the "needs-wpt" label

Patrick: so might be good to look at this...

ACTION: Olli to review #509 for potential test

Triage unlabelled issues https://github.com/w3c/pointerevents/issues?q=is%3Aissue+is%3Aopen+-label%3Afuture+-label%3Av3++-label%3Av3-blocking

Patrick: everything now is tagged with either "v3"/"v3-blocking" or "future" (with exception of the one issue #506 that is more administrative)

Patrick: any issues we want to look at specifically?

Olli: nothing from me

Patrick: from the list of blocking issues https://github.com/w3c/pointerevents/issues?q=is%3Aissue%20state%3Aopen%20label%3Av3-blocking

Patrick: w3c/pointerevents#504

Patrick: should just be a change to prose, nothing substantive

Mustaq: I can send a pull request next week

ACTION: Mustaq to submit PR for #504

Patrick: w3c/pointerevents#513

Olli: that's the one I promised a PR for but didn't have time due to vacation

Patrick: no problem

ACTION: Olli to look at #513

Patrick: let's look at v3 in general https://github.com/w3c/pointerevents/issues?q=is%3Aissue%20state%3Aopen%20label%3Av3

Patrick: w3c/pointerevents#509

Patrick: sorry...revisiting the same...we discussed this earlier

Olli: there's this one that Mustaq reopened w3c/pointerevents#457

[discussion on synthetic pointermove vs mousemove - should pointermove events be faked?]

Mustaq: should pointer events not send fake/synthetic events, even though in past authors may have relied on synthetic mouse events?

Olli: yes, that would be my preference, as it keeps pointer events nice and clear - you only get it when it actually moves

<smaug> https://wpt.fyi/results/pointerevents/pointerevent_pointerout_no_pointer_movement.html?label=master&label=experimental&aligned&view=interop&q=label%3Ainterop-2023-events

Olli: this is failing only in chromium

<smaug> https://wpt.fyi/results/uievents/mouse/layout_change_should_fire_mouseover.html?label=master&label=experimental&aligned

Olli: this hints that webkit doesn't send any synthetic events, ever

Mustaq: I'd conclude the same

Olli: so do we want more consistency betwen mouse and pointer events, or keep PE clearer and better defined

Mustaq: need to check how to handle hover effect

Mustaq: should close issue #459 as there's too many concerns/discussions, and refocus on more specific issues

successpr tp/ACTION: Mustaq to file new issue as follow-up/successor to #549/successpr tp

ACTION: Mustaq to file new issue as follow-up/successor to #549

Mustaq: would we want this v3-blocking?

Patrick: depends how easy/hard it is. if it's only a small wording tweak and a test, might be good to get it in for v3

Mustaq: or should we continue on the same issue?

Olli: probably good to have one specifically about synthetic events in general

Patrick: right, I think that covers all v3/v3-blocking things we need to discuss at this point. We can break early. Thank you as ever, we'll reconvene in 2 weeks' time

Summary of action items

  1. Patrick to draft PR for #517
  2. Olli to review #509 for potential test
  3. Mustaq to submit PR for #504
  4. Olli to look at #513
  5. Mustaq to file new issue as follow-up/successor to #549
Minutes manually created (not a transcript), formatted by scribe.perl version 238 (Fri Oct 18 20:51:13 2024 UTC).

Diagnostics

Warning: ‘s/ACTION: Mustaq to file new issue as follow-up/successpr tp/ACTION: Mustaq to file new issue as follow-up/successor to #549/’ interpreted as replacing ‘ACTION: Mustaq to file new issue as follow-up’ by ‘successpr tp/ACTION: Mustaq to file new issue as follow-up/successor to #549’

Succeeded: s/ACTION: Mustaq to file new issue as follow-up/successpr tp/ACTION: Mustaq to file new issue as follow-up/successor to #549/

Maybe present: Olli, Patrick

All speakers: Mustaq, Olli, Patrick, PLH

Active on IRC: Patrick_H_Lauke, smaug