[systemd-devel] Rethinking getty and fast user switching

[Colin Guthrie]

'Twas brillig, and Jan Engelhardt at 21/02/12 11:27 did gyre and gimble:
>> Why have they been cultured by any other OS? I'm not making any
>> comparisons here, just trying to suggest that things work in a logical
>> manner rather than carrying historic cruft that really has no place now.
> 
> The status quo is quite logical to me, but alas I reckon expecting
> everyone to see logic on a Vulcan-like level is asking too much.
> 
> Take f.i.: vt1 X, vt2 X, vt3 X, vt4 text, vt5 X, vt6 text, vt7 X, vt8
> text, vt9 text, vt10 text, vt11 X. Logical? Absolutely (w.r.t. some
> pattern). To you? Probably absolutely not. (Solution: VT numbers
> which are prime get an X server, non-primes get text.)

Ahh an excellent layout and one I very much like! I would also approve
of X only being started on vt #'s that feature in the Fibonacci sequence.

However, with the current status quo, this is not how things are setup
(to the best of my knowledge anyway). It's more or less a first come,
first served basis just now. So you'd have to spend some time after boot
starting Xs and swithing VTs at the right time to configure your perfect
primes layout.

However with the "appropriate login" scheme I describe you could set
things up such that just switching to those vts would start the
appropriate thing - X+dm or getty. That way your primes layout is
preserved and not left as a task for the user to configure after every
boot... (provided you booted to graphical.target of course - otherwise
they'd all be text).

:)

Col




-- 

Colin Guthrie
gmane(at)colin.guthr.ie
http://colin.guthr.ie/

Day Job:
  Tribalogic Limited http://www.tribalogic.net/
Open Source:
  Mageia Contributor http://www.mageia.org/
  PulseAudio Hacker http://www.pulseaudio.org/
  Trac Hacker http://trac.edgewall.org/