Minutes from PEWG meeting 11 March 2026

Dear all,

the minutes from today's meeting are at 
https://www.w3.org/2026/03/11-pointerevents-minutes.html and copied below:


PEWG
11 March 2026

Agenda: 
https://www.w3.org/events/meetings/bc0bed33-fd93-40a6-95bb-10f27c641863/20260311T100000/
IRC log: https://www.w3.org/2026/03/11-pointerevents-irc

Attendees
flackr, mustaq, Patrick_H_Lauke, plh, smaug, vmpstr

Chair: Patrick H. Lauke
Scribe: Patrick H. Lauke


* W3C Breakout Days 2026 w3c/breakouts-day-2026#3
* Expose 'inertial scrolling state' in wheel events w3c/pointerevents#587


# W3C Breakout Days 2026 w3c/breakouts-day-2026#3

Patrick: mentioned in a previous call that i had an email from matthew 
atkinson. specifically this was about longpress. suggested to him to 
join the breakout day for discussion

Patrick: longpress does feel like ... a strange one. touches on many 
aspects of when does a UA actually recognise something's a longpress

Patrick: also, most UAs already DO something in response. like 
iOS/Android bring up a context menu so you can catch it from that

Patrick: will forward his email to the list though


# Expose 'inertial scrolling state' in wheel events w3c/pointerevents#587

Vlad: Proposal to expose momentum scrolling in wheel events

flackr: we don'thave wheelstart / wheelend. if we want momentum, feels 
like we should also have those. but for minimal change, we would want to 
just expose the delta/that an event was fired as part of the momentum event

smaug: was briefly testing this, and we have something similar 
internally on gecko. found an oddity on linux where scrolling still 
happens but no wheel event is fired

flackr: think it's an issue in linux. think it's in the gtk layer. with 
this proposal, every browser should fire a wheel event with the flag, 
otherwise some developers now don't get wheel when it's a momentum scroll

smaug: just having a flag should be easy. we'll just have to do 
something on linux

<plh> https://docs.gtk.org/gtk4/class.EventControllerScroll.html

Patrick: so sounds like we're in agreement for the principle

flackr: are we ok with the proposed implementation?

could be a boolean, but if we want an unknown value, we can'thave that 
with boolean

<plh> proposal

as long as we can say in the spec what happens when it's unknown

flackr: i think boolean is ok...

smaug: would be annoying if later on we know that we DO want another value

fuzzy logic, plh

concrete proposal: let's make it a boolean called momentum, to indicate 
if it's a momentum phase ... thing

smaug: will check with our windows folks first, but should be fine

plh: how are we going to test this?

flackr: we'll have to add the capability to generate these in web driver?

plh: in your proposal you only have desktop. what about mobile?

flackr: but you don't have scrollwheel on touchscreen. but if you 
attached a mouse you would get it

Patrick: so what's next steps? is there a PR?

Mustaq: I can give it a try

ACTION: Mustaq to create initial PR for #587

<vmpstr> vmpstr

Patrick: any other issues?

<vmpstr> w3c/csswg-drafts#12919

vmpstr: raising an issue about overscroll and touch-action interactions, 
if folks have strong opinions on it, as a heads-up

flackr: was just looking at wheel implementation. there are cases where 
we have integers rather than boolean. could be 0 or 1 now, but then we 
can extend...

smaug: i think there was something about using enum values rather than 
strings...


# Summary of action items

* Mustaq to create initial PR for #587


-- 
Patrick H. Lauke

* https://www.splintered.co.uk/
* https://github.com/patrickhlauke
* https://flickr.com/photos/redux/
* https://mastodon.social/@patrick_h_lauke

Received on Wednesday, 11 March 2026 14:51:11 UTC