Re: Stop swimming upstream

From: Fred Kleinsorge (kleinsorge_at_star.zko.dec.com)
Date: 08/30/03 

Date: Sat, 30 Aug 2003 00:12:26 GMT

"Larry Kilgallen" <Kilgallen@SpamCop.net> wrote in message
news:9cvzPLm8xfQd@eisner.encompasserve.org...
> In article <UIs3b.3587$6z6.3290@news.cpqcorp.net>, "Fred Kleinsorge"
<my-last-name@stardotzko.dec.com> writes:
> >
> > "Ken Fairfield" <My.Full.Name@intel.com> wrote in message
> > news:3F4E500D.E4980303@intel.com...
>
> >> The standard
> >> Digital/Compaq/HP Logitech 3-button mouse works as well (in
> >> 2-button mode without intervention, or in 3-button mode with a
> >> Logitech driver downloadable from their web site).
> >>
> >
> > Another brilliant design. There is no command to query the mouse type,
and
> > no bits in the packet to do anything additional or useful (including
> > determining which byte is the first one). So the user must *explicitly*
> > load a driver for something other than the "standard" 2 button mouse.
>
> Is this fixed in the USB standard ?

Yes. Vendor, device, etc. The format and codes are fairly well defined and
full. We autodetect the LK463 as well as the japanese kb, etc. I also made
the driver table driven, and provided a mechanism for a user or third party
to load new translations and customizations for their own keyboards.

The mouse will continue to work as normal. Pretty much any mouse - as per
the core X11 standard, plus the hack for the thumbwheel.

I'll probably throw out an example for writing a extended input device (X
Input Extesion)... but to some degree this will be limited until/unless we
document how to write a 3rd party USB driver. But I'll probably at some
point have one that can handle most axis control type devices.