On Fri, Jun 14, 2002 at 12:30:47PM +0200, David Ferlier wrote:
>
> All you suggested is perfect, many thanks for taking time to answer.
>

Dunno if it is perfect - I just don't want to get your effort bogged
down in a discussion about the perfect key usage and have you chuck it
all in out of disgust for the whole process.  For what it is worth, we
already have inconsistency in our keyboard interface when it comes to
entering DDB, WTF we should be wringing our hands about the lack of
pgup/pgdown keys on some machines is beyond me.
 
> So, a kernel option will enable/disable the use of scrolling. If it is set, a wsconsctl flag will be set to zero or 1 to designate the wanted mode (zero -> combos, 1 -> hold screen, 0 by default ;-).
> 

Yes and yes.  Also, as someone else suggested, being able to set the
key combo in the kernel config would be very handy as single user mode
is one place where scroll back would be extremely good to have so you
can catch those pesky errors that whiz up the screen.

> And then what would be cool is to allow the user to specify via wsconsctl as many modifiers as he wants for mode 1, like modifier 1 = alt, modifier 2 = ctrl. So people wanting complex combos won't be stuck ;-). And naturally the last wsconsctl variable, the key which will work in mode 1 and 0 (let's make them common, so when someone wants to go from mode 0 to mode 1, he will be able to keep the keys he used before)
> 

That sounds excellent, the key (no pun intended) to this is making
sure we have the flexibility to allow people to bind what they like.
Having a sensible port dependent default will not hurt - if people
don't like the default then they can change it.

> Thanks Brett. Hope your proposal is the last one of this thread ;-)
> 

heh - not bloody likely but I stand ready with clue-bat to try and
prevent you getting bitten to death by ducks.  What we need here is a
reasoned discussion of your implementation and how is can be improved
not some gum gnashing about what keys are on which keyboards.

-- 
Brett Lymn