Draft minutes: 22 June 2016 call

Hi All,

draft minutes from the last PEWG call are available at

https://www.w3.org/2016/06/22-pointerevents-minutes.html

and copied below.

If you have any comments, corrections, etc., please reply to this email 
by 28 June. In the absence of any changes, these minutes will be 
considered approved.

W3C
- DRAFT -

Pointer Events Working Group

22 Jun 2016

See also: IRC log

Attendees

Present
Mustaq_Ahmed, patrick_h_lauke, teddink, Olli_Pettay, dtapuska, Rick_Byers
Regrets
Chair
patrick_h_lauke
Scribe
dtapuska
Contents

Topics
repo cleanup? https://github.com/w3c/pointerevents/issues/90
Support hovering button states 
https://github.com/w3c/pointerevents/issues/14
Click event while a pointer event is captured 
https://github.com/w3c/pointerevents/issues/75:
Clarify click event firing for chorded buttons 
https://github.com/w3c/pointerevents/issues/93
setPointerCapture should say something about iframes 
https://github.com/w3c/pointerevents/issues/16
Should a captured pointer send boundary events by default? 
(https://github.com/w3c/pointerevents/issues/61) linked to 
https://github.com/w3c/pointerevents/issues/39 and 
https://github.com/w3c/pointerevents/issues/8
implementation update
Should a captured pointer send boundary events by default? 
(https://github.com/w3c/pointerevents/issues/61) linked to 
https://github.com/w3c/pointerevents/issues/39 and 
https://github.com/w3c/pointerevents/issues/8
Summary of Action Items
Summary of Resolutions
<patrick_h_lauke> (just grabbing a fresh coffee, and fair warning that i 
have to dash off 10 minutes before the hour)
<patrick_h_lauke> any scribe volunteers?
who is scribbing today? I can do it if you like

<patrick_h_lauke> dtapuska you're on
<patrick_h_lauke> scribe: dtapuska
PL: Doug isn't here; first agenda item was to discuss first working draft

repo cleanup? https://github.com/w3c/pointerevents/issues/90

PL: Fairly safe to remove
... anyone else have any issues please shout out

<patrick_h_lauke> i'm hearing somebody asking if we can hear them...
<patrick_h_lauke> we can, but can you hear us?
<teddink> No issues - dialing back in now. My friend Skype is not 
cooperating this morning.
<teddink> Nope - can;t hear you - dialing back in.
<patrick_h_lauke> we'll wait a minute for ted to come back
<patrick_h_lauke> just shout when you're back in
Support hovering button states 
https://github.com/w3c/pointerevents/issues/14

RB: only outstanding issue on this one is on teddink;

teddink: I think I have an answer; just give me a minute to get it into 
github
... was definitely get an answer for the anniversary edition of edge 
coming out. We treat pen as a mouse
... so that assumption is the same as we were discussing on the github issue

MA: When you are treating pen as a mouse; when the pen is touching the 
surface it is a left click dragging mouse

TD: it is a little more complicated on the windows side; as soon as it 
interacts with an app it is a pen; but when it interacts with a view 
then it is mouse

DF: Sorry to interrupt; I don't have much time; was anything needed from me?

PL: Just discussion about first working draft; we will sync up later
... So issue 14 can be closed; so treating the pen as a mouse is ok

RB: We had a proposed changed didn't we?

MA: I think we agreed we don't need a change in the spec just a test
... we are keeping the definition of click the same;

RB: We are talking about issue 14; not that one
... Scott gave us description two weeks ago.. there are some minor 
changes in the spec test
... someone should own the spec changes.

MA: I'll take it

<patrick_h_lauke> ACTION: mustaq to file pull request [recorded in 
http://www.w3.org/2016/06/22-pointerevents-minutes.html#action01]
Click event while a pointer event is captured 
https://github.com/w3c/pointerevents/issues/75:

PL: Are we ok to close issue 75?
... current spec test is fine; edge is tracking an issue

RB: Yup; we just need a web platform test
... mustaq opened an issue in web-platform-tests so we can effectively 
close this issue

<patrick_h_lauke> ACTION: close issue [recorded in 
http://www.w3.org/2016/06/22-pointerevents-minutes.html#action02]
Clarify click event firing for chorded buttons 
https://github.com/w3c/pointerevents/issues/93

MA: PE spec doesn't override the click defintion
... should fire on mouse down/up; but it isn't clear about what fires 
for corded buttons.
... Any activity for a primary button should fire the click event

RB: perhaps teddink can test it.
... this is a bit of an edge case; I don't think it matters too much
... but the spec should have a note
... I'm ok noting it either way edge implements it

TD: Ok I will look at it soon

<patrick_h_lauke> ACTION: ted to check edge implementation and comment 
on github [recorded in 
http://www.w3.org/2016/06/22-pointerevents-minutes.html#action03]
<scribe> ACTION: 93 gets assigned to teddink [recorded in 
http://www.w3.org/2016/06/22-pointerevents-minutes.html#action04]
setPointerCapture should say something about iframes 
https://github.com/w3c/pointerevents/issues/16

RB: So we have just been discussing with the security team this morning
... would be ok with a limitation with a cross origin iframe and there 
is no pointer interaction
... whether setPointerCapture isn't allowed
... sounded like jacob was ok with it but wasn't really required.
... We need to do it for sandboxed iframes; so we'd have to do the work 
anyways.
... teddink can you follow up with jacob to comment on the bug

<patrick_h_lauke> ACTION: ted to check with jacob if there are further 
issues/problems or if we should accept this [recorded in 
http://www.w3.org/2016/06/22-pointerevents-minutes.html#action05]
Should a captured pointer send boundary events by default? 
(https://github.com/w3c/pointerevents/issues/61) linked to 
https://github.com/w3c/pointerevents/issues/39 and 
https://github.com/w3c/pointerevents/issues/8

implementation update

MA: We don't have an big implementation issues; so we will move forward 
with marking feature as experimental
... one issue with maps; but they will be landing a fix to the front end 
soon

RB: Our main blocker is getting resolved with PE; we were waiting for 
maps front end to be fixed
... they have a backwards implementation checking for compatibility.. 
they were testing the touch event constructor and then testing for PE events
... but still having the pointer event handlers and canceling the events
... relatively easy to fix in maps; if you delete the touch event 
constructor pointer events works fine with maps
... so we aren't aware of any other blocking issues. So we can roll it 
out to a larger experimental audience

PL: Is it something that non-normative added to the spec?
... or is that a working group note; some pseudo code how to detect 
pointer events, touch..

RB: Ya it would be good that specs indicate about feature detection
... just trying to find the msdn page

<rbyers> https://msdn.microsoft.com/en-us/library/dn433244(v=vs.85).aspx
<smaug> is the audio bad just for me?
smaug: its fine for us

<smaug> now it is fine
PL: something might be helpful; we don't mention that certain UAs 
support PE and other things like touch events
... I can take an initial look at this; and we can hash it out in github

RB: I'll file an issue and assign it to you

<patrick_h_lauke> 
https://patrickhlauke.github.io/touch/3d-jquery-rotator/index-pointerevents.html
PL: I've been playing with PE in the dev edition. Did hit some weirdness

PE: I didn't manage to get it working; it doesn't seem to keep the swiping
... fairly sure I've added touch action; not sure how to track it down 
further

RB: I see it working with mouse and not touch

PL: does work with touch on my surface

<patrick_h_lauke> 
https://patrickhlauke.github.io/touch/tracker/tracker-width-height.html
PL: something similar I'm seeing the first movement but nothing else

RB: Just for touch
... this is likely the issue we found yesterday
... touch-action isn't always applied

teddink: touch-action only applies inside the box

RB: Can you file an issue on this; and we should block on this

<scribe> ACTION: MA will open a chromium bug [recorded in 
http://www.w3.org/2016/06/22-pointerevents-minutes.html#action06]
OP: I haven't been able to sync up with them; there are some new people 
working on itt

Should a captured pointer send boundary events by default? 
(https://github.com/w3c/pointerevents/issues/61) linked to 
https://github.com/w3c/pointerevents/issues/39 and 
https://github.com/w3c/pointerevents/issues/8

RB: we were debating about what was most logical; scott wrote some demos
... Is this waiting on a response from you teddink?

TD: Ya I was exchanging some emails with jacob and it works for the 
simple example
... but as soon as we get into more complex cases; with translation; 
svg, etc... it doesn't really work with the bounding box example
... trickier and potentially impossible when it is just a box

RB: ya the bounding box approach is optimization for the simple case
... But elementFromPoint should work for anything.

TD: Only thing that came into the case was CSS hover
... someone might have to be applying a class name
... jacob asserted that the simple case should just work with CSS

RB: Ya the simple case works purely with CSS in your design; but with 
javascript in our design
... there are pros and cons to either design. If there were actually 
websites relying on it today and that would give us concrete information
... is there any way you (@teddink) that you can provide some usage data?

TD: We don't have data that low; but we could and it would take some 
cycles to get the data

RB: I think we should try to gather our data when we come face to face 
to discuss it in Redmond

TD: Does google have any data?

RB: we don't have any data from the field because we don't have PE enabled
... other way to approach this; it is all tied up to the implicit hit 
testing..
... can you ask your team to how would you feel if chrome shipped this way
... leaving it as an open issue in the spec; and have an interop issue 
that we'd get back data from web developers

TD: 13% overall are using PE

MA: can we get the whatever list you have

TD: Ya I can send the list

RB: We both should do some searching here. We lack some experience that 
your QA team has
... does your QA team has more experience in knowing what tends to be 
issues with respect to pointer events

TD: I can certainly talk to the QA team and see if they know of 
anything. I can certainly also look at the bug database and the issues..

RB: If there are examples of sites that break when turn on touch events 
on edge desktop; those sites would certainly be important for us to look at

<patrick_h_lauke> (misheard "ship blocker" as "shit blocker"...true though)
RB: More detail the better before we meet F2F

<patrick_h_lauke> ACTION: ted to scour bug db/QA to get specific 
sites/scenarios that may be likely to expose the difference in behavior, 
work with rick on resolving this [recorded in 
http://www.w3.org/2016/06/22-pointerevents-minutes.html#action07]
<smaug> some query on the google tool on sites using setPointerCapture 
and boundary events?
RB: We will get some data from dev channel; and then the bugs will roll 
in and we can analyze

PL: we've worked through the agenda items; any other business?
... next wednesday?; see you then

Summary of Action Items

[NEW] ACTION: 93 gets assigned to teddink [recorded in 
http://www.w3.org/2016/06/22-pointerevents-minutes.html#action04]
[NEW] ACTION: close issue [recorded in 
http://www.w3.org/2016/06/22-pointerevents-minutes.html#action02]
[NEW] ACTION: MA will open a chromium bug [recorded in 
http://www.w3.org/2016/06/22-pointerevents-minutes.html#action06]
[NEW] ACTION: mustaq to file pull request [recorded in 
http://www.w3.org/2016/06/22-pointerevents-minutes.html#action01]
[NEW] ACTION: ted to check edge implementation and comment on github 
[recorded in 
http://www.w3.org/2016/06/22-pointerevents-minutes.html#action03]
[NEW] ACTION: ted to check with jacob if there are further 
issues/problems or if we should accept this [recorded in 
http://www.w3.org/2016/06/22-pointerevents-minutes.html#action05]
[NEW] ACTION: ted to scour bug db/QA to get specific sites/scenarios 
that may be likely to expose the difference in behavior, work with rick 
on resolving this [recorded in 
http://www.w3.org/2016/06/22-pointerevents-minutes.html#action07]

Summary of Resolutions

P
-- 
Patrick H. Lauke

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

Received on Thursday, 23 June 2016 10:13:35 UTC