Nathan J. Williams wrote:
> Are these really scanners that have to use the uscanner(4) interface
> instead of ugen(4)? Using ugen(4) entirely would be better, and it
> seems like problems with that (in either ugen or SANE) are what need
> to be addressed.

   Then, the question becomes "Why is there a uscanner device?".
I don't know the answer to that, but adding support for an
ioctl() to it, since it already exists, doesn't seem like
too bad of an idea...  (*shrug*)

                       - Chris