flee@gondor.psu.edu (Felix Lee) (11/16/87)
In article <1402@cuuxb.ATT.COM> dlm@cuuxb.UUCP (Dennis L. Mumaugh) writes: [tales of rm *.c instead of *.o, and modified shells] > The moral of this is that the command interpreters need to be > modified to request confirmation of potentially fatal things such > as rm * and it must be in the command interpreter as the command > itself can't know whether the list is an expanded list or a > individually entered list. Asking for confirmation is easy to overdo. Comp.risks has a few stories right now about computer operators who routinely answer "Is the correct tape mounted?" without actually checking. This is the case where a system blindly asks "What-you're-about-to-do-could- possibly-destroy-life-as-we-know-it, are-you-sure-you-want-to-do-this?" every time you light a match. What would be better is a system that follows what you're doing and says things like, "That isn't really what you meant, is it?" when you make a silly error. A system like a person looking over your shoulder, checking what you type. There are two places normal systems ask for confirmation: 1) when what you're about to do cannot be undone, like exploding a bomb; 2) when what you're about to do was probably triggered accidentally, like bumping into a fire alarm. GNU Emacs has both types of confirmation. 1) When you try to exit, it'll ask you if you want to save all your work first. 2) If you type certain keys, like ESC-ESC, Emacs pops up a window saying, "This is a highly advanced feature that will not do anything you expect, unless you know what you're doing; do you know what you're doing?" I have nothing against systems that ask for confirmation, but asking for confirmation will often just cover up design flaws (sometimes called features). Fire alarms don't ask you "Is this really a fire?" And can you imagine driving a car that asks "Are you sure you want to leave the driveway" and "Are you sure you want to crash into the tree"? In general, I prefer undoable actions. If it's undoable, you don't have to reach a firm decision immediately; you can change your mind when you like. This fits better with the idea of the computer as a tool for exploration. -- Felix Lee flee@gondor.psu.edu {cbosgd,cmcl2}!psuvax1!gondor!flee