Re: uniform input device packets?

Mathieu Bouchard (boum01@UQAH.UQuebec.CA)
Tue, 23 Jun 1998 21:01:40 -0400 (EDT)

> > Yes; USB input device support could follow fairly easily then. However, USB in
> > itself can do more than just input devices.
> Methinks this thing is starting to sound a little bit like the
> console-handling part of EvStack as proposed by the GGI folk ...
> vaguely speaking, the idea is to provide a 'unified console object'
> which delivers a stream of event messages to a process, those events
> coming from a set of input devices which are "related to" the console
> (keyboard, mouse, joystick, light pen, trackball, retinal orientation
> detector, etc). Am I wrong?

yes, it is the case.

I'll have to look at what the GGI project have done...

in the meanwhile, here is a slight rethink of the protocol:

no initial "i" byte, no terminal "\n". we're down to 14 bytes.

time signature is not uuencoded, and is always bigendian (i own a pc but
am maso). down by 2 bytes, we're at 12.

Only 2 bytes for the argument instead of three, so pure 16-bit instead of
uuencoded 4/3*18-bit; bigendian of course.

11 bytes for now... looking again at this soon, and will reply to other
mails during the night, or tomorrow.

matju

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu