at the time beeing there are quite different interpretations of the click positions across browsers.
I am stumbling upon that when catching events in inline svg with transforms on the target elements.