In message <20020416143148.J23787@uriel.eclipsed.net>, gabriel rosenkoetter wri
tes:
>
>.. though it doesn't exactly say which way those signals are going.
>(I *presume* the clock is sent by the host and received by the
>device and the data is sent by the device, but it's possible that
>data is actually two-way, as weird as that might be.)

Given that most PC keyboards have LEDs, I'd say that yes, *something* 
has to allow transmission of data from the host.

		--Steve Bellovin, http://www.research.att.com/~smb
		Full text of "Firewalls" book now at http://www.wilyhacker.com