IRC log of webevents on 2011-03-22

Timestamps are in UTC.

15:02:25 [RRSAgent]
RRSAgent has joined #webevents
15:02:25 [RRSAgent]
logging to
15:02:33 [ArtB]
RRSAgent, make log Public
15:02:57 [ArtB]
ScribeNick: ArtB
15:02:58 [ArtB]
Scribe: Art
15:02:58 [ArtB]
15:02:58 [ArtB]
Date: 22 March 2011
15:02:58 [ArtB]
Chair: Art
15:02:58 [ArtB]
Meeting: Web Events WG Voice Conference
15:03:00 [ArtB]
Regrets: Emmanuel_Nkeze
15:03:09 [Cathy]
Present+ Cathy_Chan
15:03:31 [Dzung_Tran]
Dzung_Tran has joined #webevents
15:03:40 [Dzung_Tran]
Present+ Dzung_Tran
15:04:32 [ArtB]
Present: Art_Barstow, Cathy_Chan, Josh_Soref, Matt_Brubeck, Anders_Höckersten, Olli_Pettay
15:04:43 [ArtB]
RRSAgent, make minutes
15:04:43 [RRSAgent]
I have made the request to generate ArtB
15:05:26 [smaug_]
[17:02]shepazuwill be there shortly
15:07:20 [ArtB]
Present+ Doug_Schepers
15:07:26 [ArtB]
Topic: Tweak Agenda
15:07:33 [ArtB]
AB: I posted a draft agenda yesterday ( ). The basic idea is to have explicit agenda items for the two Open Issues and then with respect to the Raised issues, get status for those with associated actions and try to determine owners/actions for the other Raised Issues. (Open and Raised: )
15:07:43 [shepazu]
Zakim, code?
15:07:43 [Zakim]
sorry, shepazu, I don't know what conference this is
15:08:05 [ArtB]
AB: Any comments or change requests?
15:08:21 [ArtB]
[ None ]
15:09:15 [ArtB]
Topic: Issue-1 Resolve touch area re. radius and angle
15:09:22 [ArtB]
AB: Issue-1 is "Resolve touch area re. radius and angle" ( ) and we discussed this issue on Feb 22 ( ).
15:09:42 [ArtB]
AB: Issue-1 has at least two associated actions: Action-16 for Doug to "Follow up with the canonical guys re copyrights" ( ) and Action-17 for Olli to "Investigate various angle-related work e.g. InkML, CSS, SVG, ..." ( )
15:10:03 [ArtB]
AB: Olli addressed Action-17 earlier this week via ( ). I think we can can consider Action-17 closed. However, Olli does raise some questions in his email.
15:10:30 [ArtB]
AB: Action-11 "Update touch events spec for next week" ( ) is somewhat generic so it's not clear if this applies specifically to this issue or if this action was created during our "tracker training session" on Feb 15.
15:11:15 [ArtB]
MB: Action-11 is related to Issue-1
15:11:23 [ArtB]
... but am waiting for other inputs too
15:11:36 [ArtB]
... I've made some other minor edits
15:11:45 [ArtB]
... but the spec needs updates to address the issue
15:12:16 [ArtB]
AB: so we'll leave action-11 open until issue-1 is resolved/closed
15:12:38 [ArtB]
AB: Olli, I think we can close action-17
15:12:40 [ArtB]
OP: yes
15:12:55 [ArtB]
... want to ask DS about radiusX and radiusY
15:13:00 [ArtB]
... is it for SVG?
15:13:07 [ArtB]
DS: yes, that's the basic rationale
15:13:14 [ArtB]
... but not really for compatibility
15:13:22 [ArtB]
... I just did some cut-and-paste there
15:13:46 [ArtB]
... I wouldn't say there is a really good reason for having those
15:14:02 [ArtB]
... and if someone has a better proposal, I'm willing to listen
15:14:34 [ArtB]
OP: if we want rX and rY we would need rotation angle to events
15:14:45 [ArtB]
... that would be close to what Canonical is doing on Linux
15:14:51 [ArtB]
... at least that is my understanding
15:15:09 [ArtB]
... would prefer degree
15:15:17 [ArtB]
DS: we don't have to be compatible with SVG
15:15:30 [ArtB]
... but it is fine if we are
15:15:52 [ArtB]
OP: does WebKit have this feature at all?
15:15:55 [ArtB]
DS: no
15:16:04 [ArtB]
OP: then do we need really need it
15:16:08 [ArtB]
DS: yes, I think so
15:16:27 [timeless_]
q+ to ask how we expect people to use this
15:16:49 [ArtB]
AB: well, any deviation from shipping deployements make it difficult to test
15:17:13 [ArtB]
DS: well, it does make it more difficult to satisfy the conformance criteria e.g. for CR
15:17:37 [ArtB]
... but that argues for us aligning with the Canonical way of doing it
15:17:46 [ArtB]
AB: well that is true
15:17:54 [ArtB]
... but they aren't really here at the table
15:18:26 [ArtB]
DS: I could consider them as an Invited Expert
15:18:42 [ArtB]
AB: would be good to get their IP commitment
15:18:48 [ArtB]
DS: ok, give me an action
15:19:21 [ArtB]
ACTION: Doug talk to Canonical about joining the WG; possibly as an Invited Expert
15:19:22 [trackbot]
Created ACTION-21 - Talk to Canonical about joining the WG; possibly as an Invited Expert [on Doug Schepers - due 2011-03-29].
15:19:41 [timeless]
15:19:59 [ArtB]
JS: would like to know how developers are going to use rotation
15:20:10 [ArtB]
... will they use that in the app
15:20:17 [ArtB]
... need to understand the expectation
15:20:42 [ArtB]
MB: for some drawing tools, rotation of touch point is important
15:21:29 [ArtB]
AB: appears we have use cases for the functionality
15:21:43 [ArtB]
... How do we move forward on this issue?
15:22:08 [ArtB]
... Is there something the Editors need from the rest of us?
15:22:26 [ArtB]
DS: I need to catch up
15:23:16 [ArtB]
MB: I am busy with FF 4 and will have more time for this spec after our release is out
15:23:34 [ArtB]
AB: is there anything else for Issue-1 for today?
15:23:38 [ArtB]
[ No ]
15:23:48 [ArtB]
Topic: Issue-7 Targets for touch events: Elements or Nodes?
15:23:51 [Cathy]
Cathy has joined #webevents
15:23:54 [ArtB]
AB: Issue-7 is "Targets for touch events: Elements or Nodes?" ( ) and it has associate Action-19 on Matt "to Address Issue-7 ( )
15:24:11 [ArtB]
AB: we discussed this on Feb 22 ( ).
15:24:32 [ArtB]
MB: I have not completed my action
15:24:42 [ArtB]
... we did have consensus the target should be Elements
15:24:55 [ArtB]
... please keep the action open and I will follow up real soon
15:25:54 [ArtB]
AB: propose this issue be resolved as Elements are the target for touch events
15:26:11 [ArtB]
DS: wondering aloud here ...
15:27:03 [ArtB]
... example: jumbled word, a letter can be grabbed
15:27:13 [ArtB]
... can isolate a piece of text
15:27:28 [ArtB]
... e.g. want 'a' of 'sad' and change it
15:27:47 [ArtB]
... can touch between the 's' and 'a'
15:27:54 [ArtB]
... How would we deal with that case?
15:28:06 [ArtB]
... Not element content
15:28:18 [ArtB]
MB: that's a hard problem
15:28:24 [ArtB]
... with mouse and other events
15:28:34 [ArtB]
... Even if use text nodes, still have granularity issues
15:28:48 [ArtB]
... would need to put each letter in its own element in the case DS described
15:28:56 [timeless]
[ you can do this with <span>s ]
15:29:01 [ArtB]
AB: so there is a way of handling that UC
15:29:11 [ArtB]
DS: yes, but it's not the best way
15:29:23 [timeless]
TextNode size is effectively random
15:29:32 [timeless]
and relates to how the parser generates them
15:29:35 [ArtB]
... There was some rationale for using Nodes
15:29:38 [timeless]
partially based on network buffering
15:29:58 [ArtB]
MB: but I don't think using Nodes will help in that case
15:30:20 [ArtB]
OP: to be able to indicate which letter is clicked, need a range object and an offset
15:30:47 [ArtB]
DS: I'm playing devil's advocate
15:31:05 [ArtB]
AB: can we live making Elements the target?
15:31:33 [ArtB]
DS: what are the advantages of making of Elements?
15:31:39 [ArtB]
OP: consistency with mouse events
15:32:09 [ArtB]
... perhaps the problem could be solved somewhere else (for touch events and mouse events)
15:32:23 [ArtB]
MB: PPK claimed early WebKit had a bug in this area
15:33:01 [mbrubeck]
15:33:02 [ArtB]
DS: I'm fine with moving forward
15:33:17 [ArtB]
... but want to make sure we agree on the reasons and document our rationale
15:34:05 [ArtB]
AB: in the absence of new info, I'd like to get agreement on this
15:34:41 [ArtB]
AB: I propose we address Issue-7 by agreeing Elements are the target for touch events (not Nodes)
15:34:48 [ArtB]
AB: any objections?
15:34:55 [ArtB]
[ None ]
15:35:38 [ArtB]
RESOLUTION: the group agrees Issue-7 should be closed with Elements being the target of touch events
15:35:54 [ArtB]
Topic: Raised Issues
15:36:02 [ArtB]
AB: we have 5 issues in the Raised state: ( ) and we had at least a brief discussion about all of them on Feb 15 ( )
15:36:23 [ArtB]
AB: I'll list them here ...
15:36:23 [ArtB]
... Issue-2 What should happen when a touch is dragged off the screen ( )
15:36:23 [ArtB]
... Issue-3 Click event target after DOM mutation during touchstart ( )
15:36:23 [ArtB]
... Issue-4 Does preventDefault on touchmove cause a dragging motion to fire a click event? ( )
15:36:25 [ArtB]
... Issue-5 What events fire if an alert is performed within a touch sequence? ( )
15:36:28 [ArtB]
... Issue-6 Touch targets in frames ( )
15:36:33 [ArtB]
AB: of those, the only one that has an open action is Issue-2 and that is Action-18 on Sangwan to "Investigate Issue-2" ( )
15:37:06 [ArtB]
AB: it would be good to identify a "owners" for these issues or proposals on what (if anything) should be done. Especially would like to see some work/proposals for those Raised Issues that have no associated actions i.e. #3, #4, #5 and #6.
15:38:04 [ArtB]
AB: we need people to commit to actively work on them
15:38:23 [ArtB]
OP: I can take Issue-5
15:39:10 [ArtB]
ACTION: olli follow-up on Issue-5
15:39:10 [trackbot]
Created ACTION-22 - Follow-up on Issue-5 [on Olli Pettay - due 2011-03-29].
15:39:29 [ArtB]
DS: I would like some other people to get active
15:39:59 [ArtB]
AB: we need someone for Issues 3, 4, 6
15:40:31 [ArtB]
DS: I'll take Issue-3
15:40:57 [ArtB]
ACTION: doug follow-up on Issue-3
15:40:58 [trackbot]
Created ACTION-23 - Follow-up on Issue-3 [on Doug Schepers - due 2011-03-29].
15:41:30 [ArtB]
AB: so now 3 of the 5 Raised issues have owners
15:41:46 [ArtB]
DS: re Issue-6
15:41:54 [ArtB]
... seems pretty straight forward
15:42:02 [ArtB]
... I think HTML5 addresses this
15:42:27 [ArtB]
... can't propagate outside the iframe (because of security)
15:43:03 [anders_hockersten]
anders_hockersten has left #webevents
15:43:08 [ArtB]
AB: here is the discussion from Feb 22:
15:43:20 [anders_hockersten]
anders_hockersten has joined #webevents
15:44:08 [ArtB]
DS: there is no question, the event should not bubble up to the parent
15:44:17 [ArtB]
OP: I don't think this is about that case
15:44:26 [ArtB]
... this is about touch start and end transaction
15:44:31 [ArtB]
... if it starts in the iframe
15:44:41 [ArtB]
... and then move finger to upper level frame
15:44:55 [ArtB]
... Does the upper level get the touch end or the lower?
15:45:17 [ArtB]
DS: I would be surprised if anyone says the parent frame should get the event
15:45:27 [ArtB]
OP: there are other tricky cases
15:45:43 [ArtB]
... f.ex start the touch and then the frame is removed
15:46:47 [ArtB]
DS: should touch events that start inside an iframe, once it is moved outside, should it propagate inside the parent?
15:46:54 [ArtB]
AB: and you say no?
15:47:12 [ArtB]
DS: for security purposes, should not get anything that was started in the iframe
15:47:46 [ArtB]
[ DS gives an example that is not minuted ... ]
15:48:41 [ArtB]
DS: there are a few options here as the touch moves outside the initial iframe ...
15:49:10 [ArtB]
... when a boundary is hit, could start new touchstart
15:50:02 [ArtB]
... there are also lots of edge cases e.g. an iframe is removed
15:50:10 [ArtB]
... or the iframes have different domains
15:50:24 [ArtB]
... There are lots of questions
15:50:59 [anders_hockersten]
it appears our phone system is not cooperating with me. I'll try to follow the rest of the discussion via irc
15:51:04 [ArtB]
ACTION: doug follow-up on Issue-6 on the email; enumerate some of the questions and sub-issues
15:51:05 [trackbot]
Created ACTION-24 - Follow-up on Issue-6 on the email; enumerate some of the questions and sub-issues [on Doug Schepers - due 2011-03-29].
15:51:39 [ArtB]
DS: seems like this should be addressed in HTML5 spec
15:51:53 [ArtB]
... but we could define this in our spec
15:52:13 [ArtB]
AB: agree we may not want to build a dependency on HTML5
15:52:29 [ArtB]
DS: yes, but, HTML5 defines iframes, security model, etc.
15:52:40 [ArtB]
... this could be coordination point for us with the HTML WG
15:52:45 [ArtB]
AB: good point;
15:53:21 [ArtB]
... after we get more discussion, whether or not we need some coordination should be clear
15:53:48 [ArtB]
AB: is there agreement this Issue-6 should be moved from Raised to Open?
15:53:53 [ArtB]
DS: certainly
15:54:07 [ArtB]
AB: does anyone think this is not an issue?
15:54:18 [ArtB]
ACTION: barstow move Issue-6 to Open state
15:54:19 [trackbot]
Created ACTION-25 - Move Issue-6 to Open state [on Arthur Barstow - due 2011-03-29].
15:54:57 [ArtB]
AB: if anyone want to help drive Issue-4 forward, please indicate that on the lit
15:55:03 [ArtB]
s/the lit/the list/
15:55:20 [ArtB]
Topic: AOB
15:55:30 [ArtB]
AB: next call March 29 (call will be one hour later again in Europe)
15:56:19 [ArtB]
... the point of reference will remain 11:00 Boston time because that is where the MIT voice conf bridge is located
15:56:31 [ArtB]
AB: anything else for today?
15:56:45 [ArtB]
DS: what is the schedule for FF4?
15:56:59 [ArtB]
MB: FF4 was release about 2 hours ago
15:57:10 [ArtB]
... and we did a mobile RC
15:57:53 [ArtB]
... note that FF4 RC2 == FF4
15:59:07 [ArtB]
AB: Meeting adjourned
15:59:18 [ArtB]
RRSAgent, make minutes
15:59:18 [RRSAgent]
I have made the request to generate ArtB
17:08:27 [ArtB]
ArtB has joined #webevents
17:24:26 [Zakim]
Zakim has left #webevents
17:33:39 [ArtB]
rrsagent, bye
17:33:39 [RRSAgent]
I see 5 open action items saved in :
17:33:39 [RRSAgent]
ACTION: Doug talk to Canonical about joining the WG; possibly as an Invited Expert [1]
17:33:39 [RRSAgent]
recorded in
17:33:39 [RRSAgent]
ACTION: olli follow-up on Issue-5 [2]
17:33:39 [RRSAgent]
recorded in
17:33:39 [RRSAgent]
ACTION: doug follow-up on Issue-3 [3]
17:33:39 [RRSAgent]
recorded in
17:33:39 [RRSAgent]
ACTION: doug follow-up on Issue-6 on the email; enumerate some of the questions and sub-issues [4]
17:33:39 [RRSAgent]
recorded in
17:33:39 [RRSAgent]
ACTION: barstow move Issue-6 to Open state [5]
17:33:39 [RRSAgent]
recorded in