begin quoting kelsey hudson as of Thu, Mar 27, 2008 at 09:25:39AM -0700:
> Lan Barnes wrote:
>
> ><grumble> but whatever happened to the *nix tradition of "return silently
> >on success"?
>
> Blame windows.
Hey, this isn't X's fault!
> people these days are trained to think that no output means something
> didn't work. Personally, I'm of the mindset that if you don't have
> something useful to say, don't say it. I trust that the program is going
> to work if I run it, because it was designed to work. If it needs to
> tell me something ("hey, I broke, and here's why!") it will.
>
> More people should expect the system to Just Work(tm) unless told
> otherwise. I don't want happy status messages filling my screen with crap.
UNIX's idea of "shut up when it's working" makes it the odd one out,
so far as OSes go (as best I know, I away correction or clarification
from the usual folks).
The fact that UNIX has effectively outlived its competitors may lead
us to forget that the UNIX way isn't the obvious-to-everyone-with-a-brain
solution...
And in a GUI world, spinners, progress meters, status messages... are
necessary, desired, and a cause for criticism if they're missing.
Perhaps we should be clamoring for a new standard stream for POSIX.
I like "stdstatus" more than "stdmisc" anyway.
--
% sh mycommand arg1 arg2 arg3 < infile > outfile 2>&1 3>&2 &
Stewart Stremler
--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
