Web Events WG Voice Conference -- 27 Sep 2011

<scribe> ScribeNick: ArtB

<scribe> Scribe: Art

Date: 27 September 2011

<ted_> zakim: nick ted_ is Ted_Mielczarek

<ssharma2> aaaa is ne

<ssharma2> Suman Sharma

Tweak Agenda

AB: I submitted a draft agenda on September 26 http://lists.w3.org/Archives/Public/public-webevents/2011JulSep/0082.html. Any change requests? I think it makes sense to combine Issue-21 and Issue-22 as one topic. We can talk about charter update during AoB.
... any change requests for the agenda?

[None]

Announcements

AB: reminder our f2f meeting at the annual TPAC meeting week is November 1 and the registration deadline is October 14 http://lists.w3.org/Archives/Public/public-webevents/2011JulSep/0079.html .
... if we do meet that day - and at the moment it is not clear if we will need to meet - it will only be in the morning (09:00-12:00 SFO time zone) and we will have a voice conference bridge for remote attendees.
... there is no requirement to come to the f2f meeting
... we have Suman joining us from Intel

SS: I work for Intel

… my group is interested in home related standards

… I attend other standards meeting

… f.ex. Khronos

DS: welcome; nice to have you on board

… it would be good to have a Khronos connection

Suman: I will definitely help as needed

AB: are you in same group as Tran?

Suman: no, he is in PC group

AB: welcome to the group!

… I have an action related to following up with Khronos so I'll contact you about that

AB: any other announcements for today?

DS: I published a new draft of the charter

<shepazu> http://www.w3.org/2010/webevents/charter/2011/Overview.html

… that includes Mouse Lock and Gamepad

AB: we will take that during AoB. Thanks!

Touch Events v1 LCWD TE

AB: reminder that October 11 is the comment deadline for the TE v1 LCWD http://www.w3.org/TR/2011/WD-touch-events-20110913/. We are doing a good job of addressing comments as they come in (i.e. not waiting until after the comment deadline).
... one administrivia issue is the LC Comment tracking document. I propose using a wiki http://lists.w3.org/Archives/Public/public-webevents/2011JulSep/0058.html . Any objections to that?

DS: why not use tracker?

AB: I find wiki easier

DS: OK

AB: I don't object to using Tracker

… but if we agree to use a wiki I'll take an action to create it and seed it

AB: any objections to using a wiki?

[None]

<scribe> ACTION: barstow create a wiki to track comments for the TE v1 LCWD [recorded in http://www.w3.org/2011/09/27-webevents-minutes.html#action01]

<trackbot> Created ACTION-76 - Create a wiki to track comments for the TE v1 LCWD [on Arthur Barstow - due 2011-10-04].

AB: Issue-19: Align initTouchEvent parameters with Webkit; any feedback from Webkit community? http://www.w3.org/2010/webevents/track/issues/19
... I note Laszlo isn't here today

… Does anyone know if there has been any related discussion by the Webkit community?

[Silence]

AB: we will continue this next meeting ...

ISSUE-23: Add a DOM4-style constructor to create and initialize TouchEvent objects

AB: Issue-23 is a result of comments from Anne http://www.w3.org/2010/webevents/track/issues/23 .
... we talked about this last week and agreed then to keep initTouchEvent method in v1 http://www.w3.org/2011/09/20-webevents-minutes.html#item03. Since then, Matt and Anne had some followups http://lists.w3.org/Archives/Public/public-webevents/2011JulSep/0077.html

<shepazu> agend+ IE10 Touch support https://blogs.msdn.com/b/ie/archive/2011/09/20/touch-input-for-ie10-and-metro-style-apps.aspx

AB: so where are we with this?

MB: Anne thinks there is no good reason to include initTouchEvent in v1

… I tend to agree with him

… I know Doug and Olli want to keep it

… I think we others to reply to Anne

… I didn't need it for my tests

DS: Matt, are your tests manual?

… or automatic

MB: they are manual

DS: previous feedback is that we want to move to automatic tests if possible

<mbrubeck> I think even if we do write automated tests, they will be very limited in what they can test.

DS: is there an analog in existing impls?

MB: no

DS: does WK have initTouchEvent today?

MB: yes, it does

… although WK's interface is different then the LCWD

… it includes some additional params

DS: and you don't think they will change their behavior?

MB: we are still waiting for feedback from the WK community

DS: I understand the approach

… but I also am concerned their is a widely implemented replacement, I have reservations about removing it

… I don't want to stand in the way

AB: would this mean TE v1 spec would have a dependency on DOM4?

DS: not necessarily

… we could just define initializer/constructor that exists in DOM4

… because DOM4 isn't likely to be done for a couple of years

… and if DOM4 then changes, we can make a revision

… I think future specs will match the more general behavior

… It would mean we need to go back to LC, I think

… Do you agree Matt?

MB: it would significantly lengthen the time to get v1 to REC

… We want to move fwd with a constructor we need for testing and will then deprecate it

DS: I think it can be used for other purposes

MB: it has been in WK since 2007 but I have seen no code in the wild that uses it

… If someone has some data shows it is being used, I'd like to see it

… ATM, only Gecko follows the spec

DS: does anyone else have an opinion?

<smaug> here we have a test using initTouchEvent :p http://mxr.mozilla.org/mozilla-central/source/content/events/test/test_bug648573.html?force=1#60

MB: if we want to take v1 fwd to match existing impls

[ Scribe missed some stuff so we pause while Matt enters his comments in IRC … ]

<mbrubeck> then we have two options: specify initTouchEvent as implemented in WebKit (with WebKit-only gesture parameters, etc.), or leave it out of v1.

<mbrubeck> If we are willing to wait longer on making v1 a Recommendation, then we have more options.

DS: it would mean we would need to back to LC (if the function is removed)

… we kinda' knew that

… as we talked about it before

… The 2nd LC would only be 3 weeks

… I think we need a discussion on the list

… Some people may object to us not having a constructor at all

… We don't want to keep bouncing back and forth

AB: I agree we need more discussion on the list

… do we need a new thread? or can we use the thread with Anne?

DS: we need to be clear about our proposal

… so we get a sense if there will be any objections

… We need to be clear on why we want to remove it

… and the implications of doing so

AB: would be good if someone could start a thread about this

… Are there any volunteers?

MB: I can respond on the ongoing thread

… I know Olli may have some feedback

… As well as others

OP: we need something for testing

MB: another question is how imp is it to finalize v1 as a REC as soon as we can?

… if we are willing to take longer, we aren't as constrained by existing impls

DS: I'd like to go REC as soon as we can

MB: why is that Doug?

DS: the patent commitments don't start until a spec reaches REC

… and that gives implementers more "confidence" re the patent risks

… but we also understand some implementers don't care about patent issues

… Some members want specs to proceed as chartered

… It would show we can make progress on something

… which is good for setting expectations

MB: ok; got it

AB: as a wrap up for today, Matt agreed to respond on the list

… is there anything else for this today?

… I think this is the most critical issue that has been so far

… So we need to think it through and get feedback

DS: removing eliminates two issues

… the Issue Laszlo has with his WK patch

… and might stop us from having to do the deprecation of initTouchEvent

… Certainly for v2 we need a more solid constructor function

… And doing the removal would get us to REC faster

Issue-21 and Issue 22

AB: Issue 21 is "Description of touchcancel event is missing some details" and it originates from one of Cathy's LC comments http://www.w3.org/2010/webevents/track/issues/21 . Matt and Cathy have related Action-72 and Action-73 http://www.w3.org/2010/webevents/track/actions/72 ; http://www.w3.org/2010/webevents/track/actions/73
... Issue 22 is "Does an element have to also register for touchstart event in order to receive touchend/touchmove events" http://www.w3.org/2010/webevents/track/issues/22 and Cathy has related Action-72 http://www.w3.org/2010/webevents/track/actions/72
... yesterday Cathy submitted proposed text to address both of these issues http://lists.w3.org/Archives/Public/public-webevents/2011JulSep/0083.html.
... the proposal is to change text in 4 sections.

MB: I haven't looked at the proposal yet

CC: there are two issues

… touchcancel description is missing some details

… to address this, I suggest a change to some existing text

… and also add a new paragraph in section 5.7 (touchcancel event)

… I also propose changing touchend and touchmove text to clarify

… (those are the 2nd and 3rd changes in my email)

AB: any comments?

… please review Cathy's proposed changes and send comments to the list

MB: at a first glance, they look good to me. Thanks Cathy!

DS: yes, thanks Cathy

AB: do we want to set a deadline for comments and if there are no comments, we consider them acceptable?

DS: yes, that's fine by me

AB: I propose then that if no one raises any issues by 12:00 Boston time on Friday Sept 30, we consider the changes acceptable

… any objections to that?

[ None ]

AB: Cathy agreed to make the changes if that's OK.

… Is that agreeable?

DS: fine with me

MB: ok with me

Testing Touch Events

AB: Olli reported on the list he hasn't done action-74 so we'll skip this topic today

Any Other Business (AOB)

AB: re adding Gamepad API and Mouse Lock API to our charter, Doug has a Draft charter that includes these two APIs http://www.w3.org/2010/webevents/charter/2011/Overview.html.
... thanks Doug!

… Did you make any changes to section 1.?

DS: yes, to add Lock and Gamepad, I needed to create some subsections

… so the Touch Interfaces is now its own section (1.1)

… I added the new specs to the Deliverables section

AB: any comments?
... this looks great

<scottmg> I had a quick read, looks good to me

… the Intentional Events is missing from the Deliverables

DS: until we get something from P&F WG, not sure we have anything to add

… I'll need to talk to them

AB: this is an interesting question

… don't think we need to block re Intentional Events

DS: I'll add it to deliverables

Suman: re games, what about depth camera?

DS: that is out of scope

… we need to be careful about adding specs in areas where there are patent concerns

… since that can prevent some Members from joining this WG

… And I would like to get more Members involved

Suman: ok; thanks for that information

… A lot of the important players are small and not W3C Members

DS: other than IP concerns, I think we should also try to keep a relatively narrow focus

Ted: yes, I agree with keeping the scope relatively narrow

DS: we also don't want to add deliverables without editors and a draft spec

… we can also recharter at some other time e.g. 6 months from now

… Let's talk about depth offline

AB: I'll respond and ask people to send comments

… what is next?

DS: I need to get some internal W3C review

… I can try to expedite the review

AB: I would like the AC review of the charter to start before the AC meeting on Nov 1

DS: I'll work toward getting an AC review as soon as I can

AB: anything else on the charter?

DS: Microsoft has implemented some touch intentional events

<smaug> https://blogs.msdn.com/b/ie/archive/2011/09/20/touch-input-for-ie10-and-metro-style-apps.aspx

<shepazu> http://msdn.microsoft.com/en-us/ie/hh272903.aspx#_DOMTouch

DS: it's unfortunate they didn't participate in our Touch Events spec

… I think there will be some interop issues

… It's simple but perhaps too simple

… They did prefix their events

… Perhaps later we can converge

AB: I think that gives us an action to followup with Microsoft
... next call is October 4, if there is sufficient topics.
... meeting adjourned

- DRAFT -

Web Events WG Voice Conference

27 Sep 2011

Attendees

Contents