From: KC5TJA <kc5tja@topaz.axisinternet.com>
  To  : ggi-develop@eskimo.com
  Date: Fri, 11 Sep 1998 11:34:29 -0700 (PDT)

Re: SYNC mode should go

On Fri, 11 Sep 1998, Marcus Sundberg wrote:
> > It still is.  It's just that it's now the responsibility of the _process_
> > to make the ioctl() to switch VCs.
> 
> Yeah right, and win 3.x had preemptive multitasking, it's just that
> it's the responsibilty of the process to switch tasks. Ugh!

Suppose YOU'RE the one who shoved the keyboard driver into raw mode,
thereby circumventing the OS's VC handler code (and it's documented as
such).  *YOU*, therefore, are responsible for switching VCs.  That's
perfectly reasonable way to do things.  You get what you pay for.  More
power -- more responsibility.  Use it wisely. 

And how, exactly, does Win3.1 fit into this?  Frankly, it doesn't.  Win3.1
doesn't have Preemptive multitasking at the Win16 API level -- it's all
cooperative, and is documented accordingly.  Only at the virtual machine
level did it preemptively multitask (and horribly at that, but still). 
And only VxDs could access the system at that level.  But you can't write
a full blown application as a VxD, so what does it matter?  Again --
there's no comparison in this argument. 

It's a matter of procuring exclusive access to a system-wide resource, and
releasing it when you're done with it.  If you can't handle the
responsibility of hardware level programming, don't do it!  But don't
complain to me about it.  I, for one, fully agree with the policies as
currently set up in the Linux kernel, with one exception -- there should
be a third mode of operation where you still get hardware events, but it's
filtered for VC switching.  But do *NOT* take away MY right to directly
access the hardware.

For what it's worth, I wrote and debugged my operating system's keyboard
driver under Linux using RAW keyboard mode.

==========================================================================
      KC5TJA/6     |                  -| TEAM DOLPHIN |-
        DM13       |                  Samuel A. Falvo II
    QRP-L #1447    |          http://www.dolphin.openprojects.net
   Oceanside, CA   |......................................................