Jon Baker wrote:

> The interface keyboards is quite strange and I dont know how to
> support an I/O line which can be either an input or an output at the
> choice of the keyboard.

  Actually, it's in effect a variant of I2C, or vice-versa (that is to
say, I2C is a variant of the PC keyboard protocol!).  Almost all rules
are the same, you have a clock and data; lines are driven by open-
collector drivers with a pull-up at one end (at least), the slave
can stretch data by holding the clock, other things elude me.

  If therefore you can write I2C code, you are off to a flying start for
a PC keyboard.  I fancy the Apple (AKB) is virtually the same concept
too (any comments?).

  Cheers,
        Paul B.