Re: RFC: A generic pointer protocol

Mark Cooke (
Thu, 14 Nov 1996 16:15:37 +0000 (GMT)

On Thu, 14 Nov 1996 wrote:

> >> Successive motion events should be clustered whenever possible: an
> >> application reading the pointing device wants the `current' information,
> > This is not always true, if one uses a drawing app (ala Photoshop)
> > then the device movement history is VERY important, [...]

Clustering is not useful in the case of, for instance, handwriting
recognition (we all want embedded Linux PDAs now don't we :) ). I would
suggest that rather than a compile time option, that an interface to
kmouse be implemented to toggle clustering.

Perhaps even some heuristic to cluster upto N events, but no more, with
rules to force locations where buttons were pressed to never be

Best Regards,


Mark Cooke The views expressed above are mine
Systems Programmer and do not reflect in any way the
University Of Birmingham current policy of my employers.