In message <m11HFNa-000g7OC@most.weird.com> Greg A. Woods writes:


>this is where the default screen emultation comes into play though --
>the character sequence generated by a *function* key should match that
>expected by the terminal

So you you're saying you do agree that, if wscons is in vt100/vt220
emulation, it *should* generate 0x7f if the big erase-leftward key
is  pressed?  Or do you not think that erasure is a function?
Is it just me, or is Greg's argument not self-consistent?

I dont really have much of an axe to grind here.  I accept that some
people want to wedge a windowing system into the kernel, despite
having another one in user-space. I will buy the kernel bloat that
causes.  Given all that, I think least-surprise is a good underlying
principle.  

Problem is, what constitutes `least-surprise' is not self-evident:
some of us think one thing, some think another. Nobody is going to win
just by asserting that what fits their biases is obviously the `right'
answer.