W3C

– DRAFT –
PEWG

29 July 2026

Attendees

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

Meeting minutes

<smaug> just a sec

harmonisation of spec with new added mouse and wheel events

Patrick: last time we met, we continued work/discussion of how to integrate the mouse and wheel things better, since they use different format

Patrick: from the minutes last time https://www.w3.org/2026/07/01-pointerevents-minutes.html#ActionSummary had an initial look with Olli about what parts of the tables seem redundant

Patrick: working on branch here https://github.com/w3c/pointerevents/tree/patrickhlauke-harmonise-with-ui-events

Patrick: outlining my initial big work ... the biggest part is I tried what it would look like moving the mouse and wheel sections to the end, keeping the bulk of the spec to be pointer events. thoughts?

Rob: we could also do with Examples for mouse (and wheel). effectively this spec has now morphed into the mouse events and pointer events spec... but in principle, not opposed to this

Olli: yeah i like it as well, as it highlights that PE are newer

Rob: I could also conceive that in future, pointer events will include/subsume wheel events (same way devs should use PE instead of mouse)

plh: note that in the introduction, we still have a link to UIEvents

Patrick: doh, will fix in my branch too

plh: you could also add something to the intro to "Mouse Events and Interfaces" saying that authors should consider using PE

Patrick: yeah, probably also good to add some context there... "PE are built ontop of mouse events, that's why this is here...but for new projects authors should consider using PE directly..."

<smaug> https://mozilla.pettay.fi/composed-events-dispatch.html

Patrick: other aspect i did in my draft version: i ripped out some of the rows (like sync/async, default action, composed) from the tables for mouse/wheel events.

Patrick: question is...have i ripped out too much? not enough? and once we settle on a good table format, i think we discussed introducing the same table for the PE, as it would make it more immediately skimmable for developers and implementers

Olli: things like coordinates...always the same, so repetitious stuff,and we already clarify that in MouseEvent itself

Rob: the sync/async ... i think all events are sync. only scroll event is async?

Olli: isn't it sync?

Rob: it's a posted task...

Olli: it's not part of PE

Rob: sure, but just as example that devs should not expect it right away/to be sync

Olli: for things like bubbles and cancelable, in future we may have an algo that defines these thing...but then, it might still be nice to have a quick glance

Olli: composed is something i often need to refer back to

Patrick: what about all other things in context? can rip out screenX, layerX, etc. what about UIEvent.view, Event.target...

Olli: only UIEvent.detail, button, buttons, and relatedTarget are the only ones we need to keep there...

Patrick: but even those are defined already in the interface definition... so really just UIEvent.details ? and cases where things are weird.

<plh> w3c/uievents#413

Patrick: actually, weird question... mouse and wheel in their definition still point to UIEvent. so will that spec still exist? and just define the high-level concept of a UIEvent? or will it move to DOM spec, or HTML spec, or whatever?

Olli: similar with target... (see above PR)

plh: does this affect some of our algos?

Rob: you could end up with weird questions like "can i pointerlock a pseudo element" (?)

plh: we may need to open an issue - "does it affect any places in our spec"

Rob: pointer capture, we might need to change something there

ACTION: Rob to open an issue about changes needed to our algos with regards to UI Events changes (target / pseudo target?)

plh: and pointerlock spec might also need to ask themselves that question

ACTION: Patrick continue process of simplifying/chopping tables from mouse/wheel - once we get to a nice workable table there, start porting the same type of table to pointer events

Summary of action items

  1. Rob to open an issue about changes needed to our algos with regards to UI Events changes (target / pseudo target?)
  2. Patrick continue process of simplifying/chopping tables from mouse/wheel - once we get to a nice workable table there, start porting the same type of table to pointer events
Minutes manually created (not a transcript), formatted by scribe.perl version 248 (Mon Oct 27 20:04:16 2025 UTC).

Diagnostics

No scribenick or scribe found. Guessed: Patrick_H_Lauke

Maybe present: Olli, Patrick, Rob

All speakers: Olli, Patrick, plh, Rob

Active on IRC: Patrick_H_Lauke, plh, smaug