Conversation with #webevents at Tue Apr 12 10:52:13 2011 on ArtB@irc.w3.org (irc)

Web Events WG Teleconf

Logistics

Date: 12 April 2011
Agenda: http://lists.w3.org/Archives/Public/public-webevents/2011AprJun/0033.html
Present: Art_Barstow, Cathy_Chan, Josh_Soref, Matt_Brubeck, Doug_Schepers, Laszlo_Gombos
Regrets: Sangwhan_Moon
Chair: Art
Scribe: Art

Tweak Agenda

(11:03:33) Scribe: AB: I submitted a draft agenda to the list yesterday ( http://lists.w3.org/Archives/Public/public-webevents/2011AprJun/0033.html ). It is very full and covers all of the Open, Raised and Pending Review Issues. Don't expect to talk about every issue today; for example, we probably won't have time for item #9 which lists those Issues blocked by Open Actions.
(11:04:13) Scribe: AB: any agenda change requests?
(11:04:36) Scribe: DS: are we going to talk about test today?
(11:04:42) Scribe: AB: no, but we can take that next week
(11:04:46) Scribe: DS: ok, that's fine
(11:05:12) Scribe: JS: I'm going to need to send regrets for the next 2 weeks

Issue-1 - Resolve touch area re. radius and angle

(11:05:35) Scribe: AB: Issue-1 is in the Pending Review ( http://www.w3.org/2010/webevents/track/issues/1 ). state. When we discussed this issue during our last call (March 29), we said Matt's proposal would be accepted if there were no concerns raised ( http://www.w3.org/2011/03/29-webevents-minutes.html#item02 ).
(11:05:36) mbrubeck: I will probably have to send regrets next week.
(11:06:40) Scribe: AB: Cathy started a related "Comments on rotationAngle" thread on 29 March ( http://lists.w3.org/Archives/Public/public-webevents/2011JanMar/0092.html )
(11:06:57) Scribe: AB: Matt Brubeck announced a patch on 6 April ( http://lists.w3.org/Archives/Public/public-webevents/2011AprJun/0032.html ); patch: http://dvcs.w3.org/hg/webevents/rev/348e35a8557f
(11:07:19) Scribe: AB: where do we stand on this issue?
(11:07:27) ***Cathy "old" + "beta" = pain
(11:07:45) Scribe: JS: minor request to Editors - would like to limit lines to 80 chars
(11:07:50) Scribe: MB: yes, I can do that
(11:07:55) Scribe: DS: not sure I understand
(11:08:01) Scribe: ... why?
(11:08:13) Scribe: MB: need to do a lot of scrolling with some tools
(11:08:21) ***timeless [ http://dvcs.w3.org/hg/webevents/rev/1ea45991d3e9 ]
(11:09:55) timeless: Is "(-90, 90]" well defined?
(11:09:59) Scribe: MB: some edge cases need to be addressed
(11:10:17) timeless: i know there are mathematical notations for it, but I suspect it isn't defined in any form we reference
(11:10:34) Scribe: CC: I'm fine with removing the req that rot angle be 0 when touch area is a circle
(11:10:42) Scribe: ... it is irrelevant then
(11:10:57) Scribe: ... typically want 0 but there could cases for something else
(11:11:45) Scribe: DS: this should be spelled out in the spec
(11:11:50) Scribe: ... need this to be clear
(11:12:06) timeless: sounds good to me
(11:12:15) Scribe: ... want implementors who didn't participate in this conversation to understand
(11:12:26) Scribe: AB: who can write that text?
(11:12:30) Scribe: MB: I can do it
(11:12:49) Scribe: ACTION: brubeck write explanatory text re rot angle
(11:12:49) ***trackbot noticed an ACTION. Trying to create it.
(11:12:50) trackbot: Created ACTION-31 - Write explanatory text re rot angle [on Matt Brubeck - due 2011-04-19].
(11:13:00) Scribe: JS: not sure the syntax used is explained
(11:13:09) Scribe: MB: I can define the syntax or add a ref
(11:13:41) Scribe: AB: after Matt closes this actions, are there any other sub-issues for Issue-1?
(11:13:48) Scribe: MB: don't know of any other sub-issues
(11:13:51) Scribe: JS: me neither
(11:14:12) Scribe: AB: we have a provisional resolution for Issue-1 and that is the closure of ACTION-31
(11:14:55) Scribe: ... I would like to close the issue after Matt completes action-31
(11:15:01) Scribe: ... Any objections to that?
(11:15:04) Scribe: [ None ]
(11:15:16) Scribe: ACTION: barstow close Issue-1 after Matt completes Action-31
(11:15:16) ***trackbot noticed an ACTION. Trying to create it.
(11:15:16) trackbot: Created ACTION-32 - Close Issue-1 after Matt completes Action-31 [on Arthur Barstow - due 2011-04-19].

Issue-12 - altKey/ctrlKey/metaKey/shiftKey should be attributes of TouchEvent, not TouchPoint;

(11:15:33) Scribe: AB: Issue-12 is in the Pending Review state ( http://www.w3.org/2010/webevents/track/issues/12 ). Matt included rationale to address this issue in the Issue itself and submitted a related thread on the list ( http://lists.w3.org/Archives/Public/public-webevents/2011AprJun/0017.html ).
(11:15:55) Scribe: AB: since no one responded to Matt we can take comments now; otherwise perhaps we should agree to accept the proposal.
(11:16:13) mbrubeck: I also got an (unintentionally) off-list reply from PPK in support of the proposal.
(11:16:26) Scribe: MB: the proposal has been checked in
(11:16:36) ***timeless mb: got a url for the commit?
(11:16:36) Scribe: ... moves some key modifiers
(11:16:42) Scribe: ... this matches existing impls
(11:17:09) Scribe: ... Within a single event, key mods don't differ b/w touchpoints
(11:17:14) Scribe: DS: I would think they would
(11:17:21) Scribe: MB: can you give an example?
(11:17:39) Scribe: DS: if I touch something and then add another finger
(11:18:15) Scribe: [ MB explains some more ]
(11:18:23) Scribe: DS: OK, I understand
(11:18:43) Scribe: JS: should add some text about what happens when start with 1 finger then add another and possibly another
(11:19:04) Scribe: MB: yes, I agree
(11:19:23) Scribe: AB: can someone provide related non-normative text for this?
(11:19:29) Scribe: MB: I can take an action
(11:19:48) Scribe: ACTION: brubeck add some non-normative examples about Alt Keys
(11:19:48) ***trackbot noticed an ACTION. Trying to create it.
(11:19:48) trackbot: Created ACTION-33 - Add some non-normative examples about Alt Keys [on Matt Brubeck - due 2011-04-19].
(11:19:56) mbrubeck: http://dvcs.w3.org/hg/webevents/rev/795c0f6c44e7 is the commit for ISSUE-12.
(11:20:14) Scribe: AB: other than this informative text, are there any other sub-issues here?
(11:20:18) Scribe: [ None ]
(11:20:20) ***timeless thanks ..
(11:20:47) Scribe: AB: in this case, I will close Issue-12 after Matt address Action-33. Is that OK?
(11:20:57) Scribe: ... any objections, rather?
(11:21:02) Scribe: LG: sounds good
(11:21:07) Scribe: [ No objections ]
(11:21:09) ***timeless wonders if it's possible to get hgweb diff to provide more context
(11:21:20) Scribe: AB: that's the way we will handle this

Issue-8 - initTouchEvent function

(11:21:43) Scribe: AB: Issue-8 ( http://www.w3.org/2010/webevents/track/issues/8 ) is Open per our discussion on March 29 ( http://www.w3.org/2011/03/29-webevents-minutes.html#item04 ).
(11:21:55) Scribe: AB: Matt proposed adding createTouch{Point,List} methods on March 30 ( http://lists.w3.org/Archives/Public/public-webevents/2011JanMar/0097.html ). He also posed a question about whether these methods should be added to the DocumentEvent interface in DOM 2.
(11:22:25) mbrubeck: Some Mozilla implementation work related to Issue-8: https://bugzilla.mozilla.org/show_bug.cgi?id=648573
(11:22:29) ***timeless hasn't had time to think this through ... sorry
(11:23:01) Scribe: DS: I prefer generic initializers
(11:23:05) timeless: I'd prefer generic initializers too
(11:23:13) timeless: I believe the reason for not doing it is Java
(11:23:14) Scribe: ... but some people don't like that
(11:23:20) timeless: which i believe is a stupid legacy constraint
(11:23:25) Scribe: ... they want initializers for every event
(11:23:50) Scribe: ... if possible, let's go with something generic
(11:24:09) Scribe: ... Per event initializers have issues
(11:24:26) Scribe: JS: think Java wasn't favorable to generic initializers
(11:24:51) Scribe: DS: well the arguments I heard had nothing to do with Java
(11:25:09) Scribe: ... don't think we have to worry about that concern
(11:25:30) Scribe: MB: I talked with Olli and he is doing some related implementation
(11:25:38) Scribe: ... he will send feedback to the list
(11:25:57) Scribe: AB: so, we should leave this open until we hear from Olli
(11:26:34) Scribe: ACTION: olli submit implementation feedback re Issue-8
(11:26:34) ***trackbot noticed an ACTION. Trying to create it.
(11:26:34) trackbot: Created ACTION-34 - Submit implementation feedback re Issue-8 [on Olli Pettay - due 2011-04-19].
(11:26:43) Scribe: MB: I would appreciate examples Doug
(11:26:48) Scribe: DS: yes, I can do that
(11:27:32) Scribe: MB: I think initializers are an area where having consistency with current implementations is problematic
(11:28:18) mbrubeck: I think breaking compatibility here is not a problem for adoption, because these functions are used only by test suites.
(11:28:32) Scribe: ACTION: barstow bind action-34 to Issue-8
(11:28:33) ***trackbot noticed an ACTION. Trying to create it.
(11:28:33) trackbot: Created ACTION-35 - Bind action-34 to Issue-8 [on Arthur Barstow - due 2011-04-19].
(11:29:29) Scribe: ACTION: doug Provide some initializer examples to help Matt with Issue-8
(11:29:29) ***trackbot noticed an ACTION. Trying to create it.
(11:29:29) trackbot: Created ACTION-36 - Provide some initializer examples to help Matt with Issue-8 [on Doug Schepers - due 2011-04-19].

Issue-9 - Interaction of touch events and mouse events'

(11:29:48) Scribe: AB: Issue-9 ( http://www.w3.org/2010/webevents/track/issues/9 ) is Open. We last discussed this during the 29-March-2011 call ( http://www.w3.org/2011/03/29-webevents-minutes.html#item05 )
(11:29:58) Scribe: AB: there has been some additional discussion on the list and it appears there is agreement on Matt's proposal ( http://lists.w3.org/Archives/Public/public-webevents/2011AprJun/0018.html ). At least Anders said he was OK with Matt's proposal.
(11:30:21) ***timeless thinks a simple object bag function init(params) { } ... where params = { name: "touchStart", time: new Date(), x: someX, y: someY, radius: someRadius, .... }
(11:30:28) Scribe: AB: is this ready to be closed?
(11:30:46) Scribe: MB: yes, I am ready to close this and check in my proposed changes
(11:30:50) ***timeless is trying to read
(11:30:57) Scribe: AB: does anyone object to Matt's proposal?
(11:31:08) Scribe: DS: no objection
(11:31:46) ***mbrubeck proposed adding something like "If the user agent dispatches both touch events and mouse events in response to the same user action, then the 'touchstart' event must be dispatched before any mouse events for that action."
(11:32:02) Scribe: JS: no objection
(11:32:21) Scribe: AB: proposed resolution to accept Matt's proposal for Issue-9.
(11:32:34) Scribe: RESOLUTION: group agrees with Matt's proposal for Issue-9
(11:32:41) mbrubeck: ok
(11:32:47) ***timeless thanks mb for providing that... .... had just found it, but it was hard to find browsing through the bits. the quoting was painful

Issue-10 - relatedTarget attribute;

(11:33:13) Scribe: AB: Issue-10 ( http://www.w3.org/2010/webevents/track/issues/10 ) was Raised by Matt. First, let's determine if we agree it is an issue and if so, we need someone to agree to submit a proposal.
(11:33:57) Scribe: MB: re touch enter and touch leave
(11:34:04) Scribe: ... two elements are involved
(11:34:10) timeless: Editorial, the issue description should be changed § Should "touchenter" and "touchleave" events should have the same relatedTarget attribute? § -- s/should have/have/
(11:34:15) Scribe: ... for mouse over there is a related target
(11:34:29) Scribe: ... For consistency, think the same attribute makes sense
(11:34:55) Scribe: AB: any comments?
(11:35:01) Scribe: JS: seems reasonable to me
(11:35:08) Scribe: DS: yes, need consistency
(11:35:22) ***mbrubeck fixed the typo pointed out by timeless above
(11:35:46) Scribe: AB: do you have a proposal in the issue, Matt?
(11:35:48) ***timeless :)
(11:36:22) Scribe: MB: nothing written; mostly the same as text from DOM L2 Mouse Events
(11:36:46) smaug_ [chatzilla@128.214.5.11] entered the room.
(11:36:57) Scribe: ACTION: brubeck submit a proposal to the list to address Issue-10
(11:36:58) ***trackbot noticed an ACTION. Trying to create it.
(11:36:58) trackbot: Created ACTION-37 - Submit a proposal to the list to address Issue-10 [on Matt Brubeck - due 2011-04-19].
(11:37:07) Scribe: ACTION: barstow bind issue-10 with action-37
(11:37:08) ***trackbot noticed an ACTION. Trying to create it.
(11:37:08) trackbot: Created ACTION-38 - Bind issue-10 with action-37 [on Arthur Barstow - due 2011-04-19].

Issue-13 - What event(s), if any, should fire when non-positional attributes of a touch point change?

(11:37:38) Scribe: AB: Issue-13 ( http://www.w3.org/2010/webevents/track/issues/13 ) was Raised by Cathy. Matt submitted a proposal to the list ( http://lists.w3.org/Archives/Public/public-webevents/2011AprJun/0029.html ). Let's start with a brief introduction by Matt and then open the floor for comments
(11:38:14) Scribe: CC: the text on touchmove just talks about moving touch point
(11:38:22) Scribe: ... some attrs don't have to do with movement
(11:38:42) Scribe: ... Things like force being applied to touch
(11:38:50) Scribe: DS: real good questions
(11:38:59) Scribe: ... and think touchmove is the right place
(11:39:09) Scribe: CC: think we need a single place to capture
(11:39:14) Scribe: JS: what about key strokes
(11:39:33) smaug_: Issue-13 is related to the question whether to create new touchpoint whenever something changes, or should implementation just modify the existing object
(11:39:33) Scribe: ... and don't want too many events if there is a little "jitter"
(11:39:49) Scribe: MB: for the 1st issues, do not want to fire events if only change is key mods
(11:39:57) Scribe: CC: I'm fine with that
(11:40:02) smaug_: (sorry, I can't call in, since the network connection is really bad here)
(11:40:09) Scribe: MB: as to jitter, we don't have anything now
(11:40:23) Scribe: ... not sure how they should be coalesced or filtered out
(11:41:16) Scribe: DS: not sure what spec can say other than this is impl-specific
(11:41:27) Scribe: ... then number of events fired is up to the impl
(11:42:05) Scribe: DS: I have never heard sample rate is too slow
(11:42:12) Scribe: ... not sure it is a real issue
(11:42:51) Scribe: AB: Josh, you raised this, are you OK with some non-normative text that mentions the issue and says it is impl-specific?
(11:42:54) Scribe: JS: yes
(11:43:26) Scribe: AB: is Matt's proposal complete?
(11:43:31) Scribe: CC: yes, that is fine
(11:43:48) Scribe: DS: and it would be good for sample rates to be added as impl-specific
(11:43:53) Scribe: MB: yes, I can do that
(11:44:19) ***mbrubeck action brubeck fix the spec
(11:44:50) Scribe: ACTION: brubeck commit proposal to issue-13 to Hg and add some non-normative text re sample rates being impl-specific
(11:44:50) ***trackbot noticed an ACTION. Trying to create it.
(11:44:50) trackbot: Created ACTION-39 - Commit proposal to issue-13 to Hg and add some non-normative text re sample rates being impl-specific [on Matt Brubeck - due 2011-04-19].
(11:45:41) Scribe: DS: are you using commit before review or v-v?
(11:45:45) Scribe: MB: I'm using both
(11:46:03) Scribe: DS: ok; that's fine; I'm personally OK with you just doing a commit and then review
(11:46:15) Scribe: MB: I am willing to slow down on changes if people want
(11:46:35) Scribe: DS: I don't think we want to slow Matt down
(11:46:49) Scribe: AB: I agree with Doug
(11:47:13) Scribe: ... and I also want to say that people should always feel free to raise questions and issues at any time
(11:47:20) mbrubeck: Yeah, feel free to discuss changes on the list even if they are already committed; I'm always willing to back things out and change them.
(11:47:47) Scribe: ACTION: close issue-13 after Matt addresses action-39
(11:47:47) ***trackbot noticed an ACTION. Trying to create it.
(11:47:48) trackbot: Sorry, couldn't find user - close

timeStamp attribute - proposal to remove it in the absence of compelling use case

(11:48:19) Scribe: AB: Matt started a thread about the Use Cases for the timeStamp attribute ( http://lists.w3.org/Archives/Public/public-webevents/2011JanMar/0100.html ). Anders mentioned a use case and Matt described a work around.
(11:49:27) Scribe: DS: I put it in the original ED to inherit from the base
(11:50:00) Scribe: MB: the casing caused some confusion
(11:50:36) Scribe: DS: I am OK with removing it
(11:50:41) Scribe: ... we can get it from Event
(11:51:11) Scribe: ... If we need to describe a mapping from high to/from low, may need it
(11:51:27) Scribe: MB: but if here it should inherited by TouchEvent from Event interface
(11:52:05) Scribe: DS: should be on TouchEvent and removed from TouchPoint?
(11:52:15) Scribe: MB: yes, that would be fine
(11:52:25) Scribe: ... not sure how to mark it as inherited with ReSpec
(11:52:37) ***mbrubeck likes the idea of a ToucheEvent. :)
(11:52:46) Scribe: DS: having it marked as inherited would be good; not sure how to do it
(11:53:15) Scribe: AB: do we need to have a discussion with Cameron [Editor of Web IDL]
(11:53:48) Scribe: ACTION: doug Talk to Cameron about how to use inherited with TouchEvent interface
(11:53:48) ***trackbot noticed an ACTION. Trying to create it.
(11:53:48) trackbot: Created ACTION-40 - Talk to Cameron about how to use inherited with TouchEvent interface [on Doug Schepers - due 2011-04-19].

Open/PendingReview Actions that are blocking Issue resolution:

(11:54:44) Scribe: AB: we have some Issues that are blocked pending Open or PendingReview Actions.
(11:54:50) Scribe: AB: Action-18 & Issue-2; Sangwan Moon; http://www.w3.org/2010/webevents/track/actions/18
(11:54:55) Scribe: AB: Action-22 & Issue-5; Olli Pettay; http://www.w3.org/2010/webevents/track/actions/22
(11:54:59) Scribe: AB: Action-23 & Issue-3; Doug Schepers; http://www.w3.org/2010/webevents/track/actions/23
(11:55:05) Scribe: AB: Action-24 & Issue-6; Doug Schepers; http://www.w3.org/2010/webevents/track/actions/24
(11:55:12) Scribe: AB: Action-27 & Issue-11; Laszlo Gombos; http://www.w3.org/2010/webevents/track/actions/27
(11:56:09) mbrubeck: For Action-22 / Issue-5, smaug_ was working on some test cases and found that "touchcancel" is fired in existing implementations when synchronous calls interrupt touch sequences.
(11:56:52) Scribe: AB: please follow-up on your Actions, especially if they are blocking an Issue being resolved

Testing

(11:57:08) smaug_: mbrubeck: ... if the touch transaction was ended while the sync thing was still running
(11:57:35) Scribe: DS: re the e-mail about marking up the spec such that it facilitates testing
(11:57:43) Scribe: ... the question is do we want to do that?
(11:57:53) Scribe: MB: no strong opinion; I'm fine with doing that
(11:58:00) Scribe: DS: I would like to do that
(11:58:45) Scribe: AB: I'd like to hear from the editors
(11:58:55) Scribe: ... could save us time
(11:59:12) Scribe: DS: don't know if it saves Editors time but do think it will save the group time
(12:00:01) Scribe: DS: can we use WebKit's Touch tests?
(12:00:26) Scribe: AB: based on what I've learned, they require a WebKit impl to run
(12:00:43) Scribe: DS: is anyone read to commit to writing test cases?
(12:00:58) mbrubeck: Example test: http://dvcs.w3.org/hg/webevents/file/795c0f6c44e7/test
(12:01:04) Scribe: LG: we need an implementation to run tests
(12:01:11) mbrubeck: http://dvcs.w3.org/hg/webevents/raw-file/795c0f6c44e7/test/touchevents/touch-area.html
(12:01:15) Scribe: DS: we can take this up next week
(12:01:33) Scribe: ... we can reuse the test harness being used by other WGs
(12:01:44) Scribe: ... need to think about automatic vs. manual tests

AoB

(12:02:15) Scribe: AB: next call April 26
(12:02:27) Scribe: DS: Matt has been doing a Great Job!
(12:02:42) Scribe: DS: others have sent regrets too
(12:03:21) Scribe: AB: I'm OK and will do some followups
(12:03:43) Scribe: AB: Meeting Adjourned

Actions Created

(11:12:50) trackbot: Created ACTION-31 - Write explanatory text re rot angle [on Matt Brubeck - due 2011-04-19].
(11:15:16) trackbot: Created ACTION-32 - Close Issue-1 after Matt completes Action-31 [on Arthur Barstow - due 2011-04-19].
(11:19:48) trackbot: Created ACTION-33 - Add some non-normative examples about Alt Keys [on Matt Brubeck - due 2011-04-19].
(11:26:34) trackbot: Created ACTION-34 - Submit implementation feedback re Issue-8 [on Olli Pettay - due 2011-04-19].
(11:28:33) trackbot: Created ACTION-35 - Bind action-34 to Issue-8 [on Arthur Barstow - due 2011-04-19].
(11:29:29) trackbot: Created ACTION-36 - Provide some initializer examples to help Matt with Issue-8 [on Doug Schepers - due 2011-04-19].
(11:36:58) trackbot: Created ACTION-37 - Submit a proposal to the list to address Issue-10 [on Matt Brubeck - due 2011-04-19].
(11:37:08) trackbot: Created ACTION-38 - Bind issue-10 with action-37 [on Arthur Barstow - due 2011-04-19].
(11:44:50) trackbot: Created ACTION-39 - Commit proposal to issue-13 to Hg and add some non-normative text re sample rates being impl-specific [on Matt Brubeck - due 2011-04-19].
(11:53:48) trackbot: Created ACTION-40 - Talk to Cameron about how to use inherited with TouchEvent interface [on Doug Schepers - due 2011-04-19].