Catching up on News

[Ray Butterworth]

In article <29132 at watmath.waterloo.edu>, sahayman at iuvax.cs.indiana.edu writes:
> ... 
> except at MFCF for historical reasons, the "yes" command has been
> renamed "adnauseam" (hey guys, with all this pathname reorganization,
> isn't it  about time to restore the original "/usr/ucb/yes" command.)

Don't forget what those historical reasons were.
People running interactive programs that prompt for yes or no answers
will often use type-ahead when the system is slow and they know what
the next question is going to be.  Any mistake in this can leave a
"yes" waiting to execute, and when it does an infinite stream of "y"s,
one per line, is printed.  This is really fun on a hard-copy console
when the operator returns a few minutes later and has to replace
the box of paper that was just chewed up.

Perhaps a better change would have been to make "yes"
not work if output is to a tty?