I understand. Proper error handling is hard
In Python, Java, C#, LISP and such you'd get a stack trace. That's hateful, but way better than errno 20, and works for the laziest programmer. BTW, that's one advantage of a programming language over a shitload of "small tools doing their job well" (or not so small and not so well) glued together with pipes and filters. I prefer debugging an embedded system with no OS over debugging a set of *sh scripts on any day.
Anything that causes pain when using scripts running it with #!/*bin/tcsh is good. Writing scripts in tcsh is so hateful that anything that hastens the day when hordes of angry users descend on the programmers who did it and string them up by their own guts should be heartily encouraged.
I hate software, not people. This reasoning justifies terrorism in the spirit of the worst of BOFHs. What about replacing tcsh with a program rebooting the system when not in interactive mode?
Here we have a piece of shite, named tcsh. The distributors of the shite should at least try to keep it in one piece, instead of smearing it all over the place.
