No, the DOM events are not changed by bug 605648. I don't see any difference between Nightly and Fx3.6 on the yahoo map.
Right, the DOM events are the same, but the problem is that the application is make assumptions to how much scroll delta the event represents. This previously worked in a system where one event equals one wheel notch. Have you tried yahoo maps with a logitech mouse and setpoint installed?
Oh, doesn't our mouse wheel event have delta property? http://mxr.mozilla.org/mozilla-central/source/dom/interfaces/events/nsIDOMMouseScrollEvent.idl We set the delta value to detail property of UI event... http://mxr.mozilla.org/mozilla-central/source/content/events/src/nsDOMMouseScrollEvent.cpp#59 > Have you tried yahoo maps with a logitech mouse and setpoint installed? Yes. I feel it's too fast. I guess that you mean we should implement DOM3 MouseScroll Event?
oops, DOM3 spec is here: http://dev.w3.org/2006/webapi/DOM-Level-3-Events/html/DOM3-Events.html#event-type-wheel
D3E Wheel events are pretty close to what we have.
Do we report wrong values in MouseScroll.detail, or do we send more events? Or does maps.yahoo.com not read .detail at all, but just count events (if that is the case, this would be an evangelism bug, I think). Does http://maps.google.com/ have the same problem as http://maps.yahoo.com ?
http://maps.google.com seems to invoke 'preventDefault' on the wheel event, but I'm not seeing the same behavior for http://maps.yahoo.com (perhaps because the site never has a vertical scroll bar). Could we use the 'defaultPrevented' state as a hint to aggregate the wheel events?
>Could we use the 'defaultPrevented' state as a hint to aggregate the wheel >events? I believe Logitech did exactly that in an extension to help address this problem.