[ISO-8859-1] Niklas_H�glund writes: > On Mon, 25 Sep 2000, OKUJI Yoshinori wrote: > > > We also need to decide on the external kbd interface and make > > > it official (somewhere in <mach/i386/kd.h> or so). Several programs > > > are shipping their own version already. > > > > IMO, /dev/kdb should disappear. Having two devices to read a > > keyboard is nonsense. Instead, /dev/console should support an > > interface to read raw key codes. Once Mach just exports a bare console > > device (instead of the current cooked device) to the user space, this > > would be straightforward. > > In some cases it might be good to be able to connect multiple keyboards, > but not the same number of screens. It seems more modular to me to let the > console use a keyboard device and a output device.
Indeed--I've been wishing, for a while, that I had a way to connect multiple keyboards to my PC simultaneously--for one, I could simulate a 200-key keyboard (I could have meta, hyper, super, *and* Windows keys:), along with Greek and whatever other keys I found convenient). The idea of connecting multiple keyboards just like one connects multiple joysticks (for multiplayer games or such) is a rather nice one, too. Of course, there isn't a lot of application software that supports multikeyboard input (yet)..., but just think of the whole new type of interaction between the PC and the users if a single console (or application) supported an arbitrarily-sized set of simultaneous arbitrary input devices (I can think of numerous co-operative editing sessions during which I'd rather not have had to say `Oh--give me the keyboard now'). I've already got my computer running multiple pointing devices, without any shared button functionality between them (though I have been rather disappointed that they need to share the cursor).... With USB becoming popular, it would appear that it's becoming -physically- possible to connect numerous devices of any sort to the computer..., but the software still won't let me do it.
