On Wed, Mar 04, 2009 at 06:20:54PM +0100, Daniel Mack wrote:

quoted

2) I have a patch to the driver that allows the axis to be a REL_* type.
I will post this patch for review after the driver is committed.

Cool, that might be a good alternative for users.

quoted

3) Are both 'inverted_*' flags really needed?  It appears that the end
effect of these just reverses the logical direction of the encoder.

	inverted_a	inverted_b	result
	0		0		normal encoder
	0		1		backwards encoder
	1		0		backwards encoder
	1		1		normal encoder

The driver has now been commited to input.git/next, so you could send a
patch with your ideas on top of it if you like and find some time :)

Daniel