Here's a dump from all the data I've collected on this problem so far: http://hartshorne.ca/2006/01/18/javascript_events/
There are two big problems. One is that IE's clientX/Y values seem to be 2 pixels out, from the start. Here's all I could find at MSDN:
http://msdn.microsoft.com/workshop/author/dhtml/reference/methods/ getboundingclientrect.asp
http://msdn.microsoft.com/workshop/browser/mshtml/reference/ifaces/ element2/getboundingclientrect.asp
From those pages: "In Microsoft Internet Explorer 5, the window's upper-left is at 2,2 (pixels) with respect to the true client."
So the only remaining mystery seems to be the loss of padding or border or margin in the offsetTop calculations. Can anyone point me to more information on this?
Thanks! On 18-Jan-06, at 1:17 PM, Ian MacLeod wrote:
Ah, this is good to know. I was working mainly off the research that Peter-Paul Koch was doing (both on quirksmode and other publications).
