home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.unix.wizards:4795 comp.unix.shell:4788 comp.unix.misc:4255
- Newsgroups: comp.unix.wizards,comp.unix.shell,comp.unix.misc
- Path: sparky!uunet!seas.smu.edu!utacfd.uta.edu!rwsys!sneaky!gordon
- From: gordon@sneaky.lonestar.org (Gordon Burditt)
- Subject: Re: The Problem with UNIX
- Message-ID: <Bxt5CG.1oJ@sneaky.lonestar.org>
- Organization: Gordon Burditt
- References: <1992Nov9.172715.16367@cs.wisc.edu> <aldavi01.721333614@starbase.spd.louisville.edu> <1992Nov11.194557.16258@yarc.uucp>
- Date: Mon, 16 Nov 1992 12:00:09 GMT
- Lines: 23
-
- > I think this is kinda what he is trying to determine. Typicaly UNIX
- >imposes far more structure on the user than is desired. I agree an
- >experienced user should be able to recognize and avoid the problems
- >above. What's wrong with the shell catching such obvious errors and
- >either reporting them or taking other appropriate action (ie correcting
-
- The shell has no business knowing that a program called "mail" doesn't
- take binary files as input. (For one thing, it has no way of knowing
- that the "mail" in the user's local bin directory doesn't check for binary
- files and uuencode them). It is the job of the "mail" program itself
- to check its input and either convert it or object to it.
-
- The problem with "cat a b > b" is harder. "cat" can't do the check
- because the damage has already been done. But you can't generalize
- this; something like "mail a b > b < text" DOES NOT clobber one of
- its input files even though this command line is very similar to
- the one using "cat".
-
- Now, how much does the shell have to know about every program that
- it invokes, and how does it get that information?
-
- Gordon L. Burditt
- sneaky.lonestar.org!gordon
-