6 Mar 2006

See also: IRC log


Anne_van_Kesteren, iand, chaals, gorm, christophe, sicking, ???, mjs, Stephane_Sire


Confirm minutes

<chaals> RESOLUTION: Accept minutes from face to face in Oslo

Public minutes

<chaals> RESOLUTION: CMN to publish public version.

Face to face meetings 2 and 3

<chaals> RESOLUTION: Option is open for meeting UK in july

RESOLUTION: meeting 1st-3rd may, Mountain View, we will plan on 3rd meeting in Marocco in september unless someone objects

issues list

<anne> Open issues: http://www.w3.org/2005/06/tracker/webapi/issues/open

AK: focus and blur will apply to every element but they will not bubble

JS: asked what we decided about bubbling

<bjoern> keep things as is

<bjoern> (focus/blur don't bubble)

<anne> bjoern, do you have text?

<anne> bjoern, the text for focus and blur I mean?

<bjoern> not yet, I'm on it

<chaals> ACTION: Jonas to propose text for ACTION-18 to close ISSUE-28

<trackbot> Created ACTION-89 - Propose text for ACTION-18 to close ISSUE-28 [on Jonas Sicking - due 2006-03-13].

RESOLUTION: focus and blur will be generic

who is asking about xforms?

<anne> XForms 1.0: http://www.w3.org/TR/2003/REC-xforms-20031014/slice4.html#evt-DOMFocusIn

<anne> AK: would be nice if they get errata

<chaals> ACTION: Charles to check with Xforms group about removing DOMfocusin/out and moving to focus/blur

<trackbot> Created ACTION-90 - Check with Xforms group about removing DOMfocusin/out and moving to focus/blur [on Charles Mccathienevile - due 2006-03-13].

<anne> GH: is there an issue for not having an argument in .send()?

JS: asks about no-arg send(), is there an issue?

<bjoern> They are unlikely to move to those in XForms 1.0, would probably do it for XForms 1.1; one problem is then what to say about DOMFocus; should they be deprecated in XForms?

<iand> this is action-58 -> http://www.w3.org/2005/06/tracker/webapi/actions/58

<bjoern> (we don't deprecate them, just don't define them in dom3ev...)

<iand> let's raise this as an issue

JS: let's raise officially as an issue

<chaals> RESOLUTION: Jonas to raise an issue and agree with Jim...

Mousewheel events ISSUE-30

<chaals> AvK: Lot of discussion on name...

<chaals> JS: Think Doug had some issue beyond name

<chaals> GE: Think he had an issue with wheels that are not on the mouse

JS: I'm not convinced that these non-mouse wheel devices should be "mousewheel"

<Zakim> chaals, you wanted to say that mousewheel doesn't always follow focus either

MS: I think anything that is delivered according to the mouse pointer, it should be "mousewheel", if it were delivered otherwise (e.g. according to keyboard focus) it should be a separate event

CMN: should make sure that event actually captures the intended semantics

<anne> So what is the discussion on this call about? It seems to be about two things. (1) do we want 'mousewheel' or not and (2) to what does it apply.

CMN: since "mousewheel" is already implemented by a bunch of browsers, we should stick with the name, since people know the name

JS: I know there are already other wheels

AvK: discussing two things (see above)
... could we pass a resolution that we want it, and raise an issue on what it should apply to?

Christophe: yes, people want it, especially in SVG

JS: so what are you using the event for?

Christophe: might want to do things like scroll timeline on Gantt chart, for instance

GE: have you seen wheels used for anything but scrolling and zooming?

JS: so it sounds like iPod scrollwheel and cell phone wheel should probably get the same behavior

Christophe: yes

CMN: the other use case is scrolling through menus

GE: iPod doesn't have pointer, is the wheel the same?

CMN: seems like use cases are the same kinds of things

<sicking> JS: there is an sort of orthogonal issue of where the events are dispatched. Even for mice you can sometimes configure to scroll what's under the mouse or scroll what has keyboard focus

CMN: agree that issue of where events are dispatched is kind of orthogonal
... sound of furious typing ...

<ssire> isn't it possible to define mousewheel, not assign it to a default UA action and just target it to the window object ?

JS: sounds like resolution is to have mousewheel event that applies to all types of wheel
... could be further discussion about where it was dispatched

CMN: to answer anne's question, we think we should have one, and one that applies to everything

MS: logical distinction could be made between wheel events that follow mouse vs. follow keyboard focus -- analogous to difference between mouse button events and keyboard events

CMN: in world where devices have different input, better to abstract it more

GE: my keyboard has a wheel for audio volume -- how should we handle that?

CMN: audio wheel on my keyboard only drives the audio
... if you want to get at it, we are back to issue of how to handle multiple input devices
... is there any reason not to have a mouse wheel event and call it "mousewheel"?

RESOLUTION: we have a mouse wheel event, and it's called "mousewheel"

<bjoern> someone needs an action to make a proposal for it then

CMN: are people happy that this should be a general wheel, and not just mouse wheel?
... including iPod wheels, keyboard wheels, etc

<bjoern> it should be determined by ua/platform which "wheels" it covers

SS: do you know if the device independence group has a device classification

<bjoern> we'd just say this is the kind of input device typically found on mice and similar devices

<anne> bjoern, I agree I think

I agree too

<sicking> too

<anne> bjoern, and I might be willing to take that action depending on how much work it is

CMN: question is whether we expect this to be generic

<chaals> CMN: Bjoern, right - but the question is whether we expect it to be fairly generic or very tightly bound to mice...

<anne> There is the extended interface which might make it a bit problematic.

<anne> JS: I'd argue for a generic wheel event [...]

<bjoern> devices that achieve similar effects as mouse wheels according to platform conventions / environment config

<bjoern> so yes, it'd be a bit more generic

<bjoern> it's unlikely to cover volume wheels though

CMN: some devices might change from moving pointer to wheel-like behaviour for a joystick or similar

<anne> JS: ... so mousewheel as a generic wheel event

JS: I want to clarify that I'm not arguing for another event, arguing that mousewheel is generic

<ssire> what is a wheel like behavior, is it 1D linear scroll ? or 2D linear scroll like the trackpad on the powerbook ?

CMN: sure we make a resolution?

MS: sure

JS: Doug disagreed, should we make sure to check with him?

ssire, we have a nonstandard "horizontal wheel" event in Safari


<bjoern> (when I looked into this in the SVG draft, it mentioned jog dials; what I found is that on platforms with jog dials, those are rather translated into more semantic actions and never end up as wheel rotation device events...)

<bjoern> the main problem with the event is that a script author does not know how much scrolling/zooming/whatever would be appropriate for a given .wheelDelta...

JS: there is a convention that events are named after most common device

<bjoern> (that's typically the result of os vendor adds mouse wheel event, touchpad vendor would like virtual scrolling that behaves like mouse wheel scrolling, current apps support the mouse stuff, so touchpad generates mouse wheel events...)

MS: we have a separate horizontal wheel event

<bjoern> how long before someone proposes onsteeringwheel? ...

<anne> :)

MS: we could do one of the following: add horizontalWheelDelta; have way to distinguish multiple versions of the same event; or same event struct but different event

<anne> I'd suggest .wheelDeltaX and .wheelDeltaY on the interface

CMN: could use multi-device for this

<bjoern> .wheelType = horizontal | vertical | steering | ...

JS: better to have single event with two deltas

<anne> bjoern, heh, that might work too

<anne> bjoern, good for games!

<Zakim> chaals, you wanted to ask what happens now

<bjoern> maybe ask DIWG whether they have any input

MS: we have "mousewheel" and "khtmlHorizontalWheel"
... the reason we made a separate event, instead of second wheel or separate wheel type field is that it seemed least likely to break existing apps, which expected "mousewheel" to just be a singe unidirectional wheel

RESOLUTION: let's have a mousewheel event that has a single delta, we defer on the issue of multiple mouse wheels
... we expect mousewheel to be pretty generic

<scribe> ACTION: Anne to write up text for mousewheel resolutions

<trackbot> Created ACTION-91 - Write up text for mousewheel resolutions [on Anne van Kesteren - due 2006-03-13].

<bjoern> can we resolve http://www.w3.org/mid/vgkk021fslt1kc7u36khffl831sge87t71@hive.bjoern.hoehrmann.de on TimeEvents with yes, we don't do it, action closed?

MS: can we close ISSUE-1, we passed a resolution resolving it


CMN: propose we adjourn

proposed resolution: "TimeEvents will not be adoped by DOM Level 3 Events"

<bjoern> (what to do with them instead is covered by some action on Robin to discuss this in the HCG)

<anne> we should add that we might do something with them depending on the outcome of the HCG discussion

<bjoern> would work for me

CMN: we can consider Bjoern's action item completed

<chaals> PROPOSED RESOLUTION: Bjoern's action item completed, we won't do timeEvents unless the HCG screams.

sounds good to me

<chaals> RESOLUTION: Bjoern's ACTION-13 completed, we won't do timeEvents unless the HCG screams.

<anne> ... which makes ACTION-13 completed, way to go Bjoern!

Summary of new Action Items

Minutes formatted by David Booth's scribe.perl version 1.127 (CVS log)
$Date: 2006-03-22$

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.127  of Date: 2005/08/16 15:12:03  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/face to face/face to face in Oslo/
Succeeded: s/we/Talis/
Succeeded: s/Morocco/Marocco/
Succeeded: s/JS asked/JS: asked/
Succeeded: s/focus/keyboard focus/
Succeeded: s/behavior/behaviour for a joystick or similar/
Succeeded: s/mousehweel/mousewheel/
Found ScribeNick: mjs
Inferring Scribes: mjs
Default Present: Anne_van_Kesteren, iand, chaals, gorm, christophe, sicking, +1.415.431.aaaa, mjs, Stephane_Sire
Present: Anne_van_Kesteren iand chaals gorm christophe sicking +1.415.431.aaaa mjs Stephane_Sire

WARNING: No meeting title found!
You should specify the meeting title like this:
<dbooth> Meeting: Weekly Baking Club Meeting

Got date from IRC log name: 6 Mar 2006
Guessing minutes URL: http://www.w3.org/2006/03/06-webapi-minutes.html
People with action items: anne charles jonas

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.
[End of scribe.perl diagnostic output]