> i wrote: > |n...@cs.tufts.edu wrote: > ||I am transitioning from heirloom mailx to s-mailx. I have been using > ... > ||Sample lines from my .mailrc: > || > || set replyall # doesn't work on cs > | > |Use "wysh set replyall # comment" > > But wait. Are you trying to say something, and if so, what?
I am trying to say that a comment command, instead of a lexical comment convention, is well known to be bad language design (my day job is teaching and research in programming languages), and is moreover a regression from at least prior some prior versions of Berkeley mail. I consider such a regression poor design and unjustified. The particular command that appears to the left of the would-be comment is not relevant. Consider that the POSIX shell has not only the : command but also a lexical comment convention, and it is the lexical comment that is used overwhelmingly. Consider that Tcl started out with only a comment command, but, due to outcry from users, added a lexical comment convention. I am crying out now :-) > replyall is a command that i do not like Sorry to have distracted you, but replyall is a distraction from the argument I am trying to make. The line shown is there only for the purposes of reproducing the error. The .mailrc file is a program in a language, if a simple one, and it should be treated as such. I am arguing that the language should support comments in a standard way, that a comment command is not standard and would be a bad standard, and that I can see no argument from usability that would justify it. Sorry for the strong words, but after 31 years working on language design and implementation, I feel strongly about such matters! Norman
