They did gather information from the users before implementing this; perhaps a sample from your camera wasn't submitted and it handles things differently. See http://trac.yorba.org/wiki/AVISamples