Minutes from Pointer Events WG call 4 August 2021

Dear all,

minutes from today's call are at 
https://www.w3.org/2021/08/04-pointerevents-minutes.html and copied below:

PEWG
04 August 2021

Agenda: 
https://lists.w3.org/Archives/Public/public-pointer-events/2021JulSep/0066.html

Present: flackr, mustaq, patrick_h_lauke, plh, smaug

Chair: Patrick H. Lauke
Scribe: patrick_h_lauke

* Expand note about preventing compatibility mouse events 
https://github.com/w3c/pointerevents/pull/403

* Should "click", "dblclick" and "contextmenu" events be PointerEvents? 
https://github.com/w3c/pointerevents/issues/100

TOPIC: Expand note about preventing compatibility mouse events 
https://github.com/w3c/pointerevents/pull/403

Patrick: setting scene, this is out of me being surprised by passive 
events not being able to prevent mouse compat. agree that it's kind of 
repeating things if you know DOM events, but maybe we can strike balance 
of just mentioning it for noobs like me who are not as familiar

<smaug> I guess you can't hear me

<smaug> let me rejoin

<plh> /me would like to come back to 
https://github.com/w3c/pointerevents/issues/384 if time allows

https://pr-preview.s3.amazonaws.com/w3c/pointerevents/pull/403.html#compatibility-mapping-with-mouse-events

Patrick: i'll add extra bit to the sentence to make it clear those two 
links point to [DOM] spec, and then we happy to merge?

[group agrees]

ACTION: tweak the sentence to add [DOM], then merge


TOPIC: * Should "click", "dblclick" and "contextmenu" events be 
PointerEvents? https://github.com/w3c/pointerevents/issues/100
PAtrick: (explains issue history) made a first draft 
https://github.com/w3c/pointerevents/pull/404

Mustaq: we don't have anything normative in PE at the moment about CSSOM

Patrick: we do... 
https://w3c.github.io/pointerevents/#pointerevent-interface ... ah no 
you're right, it's in a note, so non-normative

Olli: we do need something normative...

Mustaq: do you mean we normatively need to define the exception, or the 
fact that regular PE have fractional coordinates?

Olli: there's no compat problem whether browsers use fractional or not 
for regular PEs, it's only in this case with the 
click/auxclick/contextmenu that we do need to normatively make a 
statement for webcompat reasons

Patrick: if a UA doesn't use CSSOM fractional coords for regular pointer 
events (as it's not normative currently), then they would not hit the 
problem for click etc either, or rather they'd automatically satisfy this

Olli: [paraphrasing] but we do need to specify normatively for click etc 
for web compat

<mustaq> Olli's suggestion sounds good to me.

<flackr> https://w3c.github.io/pointerevents/#pointerevent-interface

https://github.com/w3c/pointerevents/pull/404/files

[discussion on how to make the requirement normative]

Patrick: could remove the first para in that note, keep the second para 
but remove the note markup around it, so that second para remains there 
and is normative

Mustaq: suggest expanding the original note at end of 4.1 to say ... 
"except for click/auxclick/contextmenu and then link to this section"

Patrick: SHOULD or MUST? this may change in future, developers may even 
moan in future about "to get fractional, i have to listen to pointerup 
rather than click"?

Robert: i think SHOULD, but can live with MUST

Olli: I think it should be MUST

<mustaq> I am on the fence!

Patrick: ok, we lean towards MUST then, can change in future if UIEVENTS 
change (the CSSOM "patch" gets applied)

Patrick: mustaq can you live with MUST

Mustaq: yes

Patrick: and fundamentally it's for UIEVENTS to grapple with this, and 
once they do, we can change our spec

Patrick: other point: should mouse compat mention this too? (thinks 
about this some more) ah no because if a UA decides to fire these, they 
are intrinsically MouseEvents, so no "conversion" or change needed

Patrick: one structural thin I noticed: when referencing 
{{PointerEvent}} in the markup, respec links to 
https://w3c.github.io/pointerevents/#dom-pointerevent

PAtrick: wondering if instead it should point to interface definition 
https://w3c.github.io/pointerevents/#pointerevent-interface

<mustaq> {{}} definition in Bikeshed here: 
https://tabatkins.github.io/bikeshed/#autolink-shortcuts

<mustaq> Better link: https://tabatkins.github.io/bikeshed/#autolinking

looking at the link to MouseEvent for instance in the note at end of 
4.1, it goes to https://www.w3.org/TR/uievents/#mouseevent

[discussion about pros/cons]

Patrick: I might make the change in a separate self-contained PR, so we 
can back it out if it does cause issues/confusion

ACTION: change the PR to make second paragraph normative, expand 
original note at end of 4.1

Patrick: looking at time, probably not worth starting new topic. would 
say we end it early now, and catch up again in 2 weeks' time. thanks all



-- 
Patrick H. Lauke

https://www.splintered.co.uk/ | https://github.com/patrickhlauke
https://flickr.com/photos/redux/ | https://www.deviantart.com/redux
twitter: @patrick_h_lauke | skype: patrick_h_lauke

Received on Wednesday, 4 August 2021 15:56:51 UTC