On Tue, 17 Jan 2006, James Youngman wrote: > > b) Even advanced users do not know of LC_COLLATE. I myself use UNIX > > operating systems for more than 10 years and it is the very first time > > I get in contact with this special environment variable. It is very > > unintuitive to set environment variables, whereas a program option is > > the normal way. Having both ways does no harm. > > I disagree; adding the option means that there are now two ways to do > the same thing: > > - one portable way (with LC_COLLATE) > > - one non-portable way > > Hence people will be using a non-portable alternative whena perfectly > portable mechanism already existed. I see this as a retrograde step. > Adding another couple of options also increases the maintenance > burden. While there is no clear benefit to the change it seems > inadvisable to make the software more complex to maintain.
So please tell me a way to use the sort program in contexts which allow no environment variable settings. I will be happy to accept it. Clean design is always good, but it should not reduce the usability. I have a problem, which is not solvable for now. I don't care for this only for theoretical reasons, but have a pratical problem. > Also, just as you didn't know about the LC_COLLATE option, the rest of > the world will not know about the new options, either, so these don't > seem to present much of an advantage. Well, options are usually documented in the manual pages. Also the documentation can contain any additional notes about portability issues when using the option (or not using it). And BTW the docs could also have a note about LC_COLLATE changing sort ordering totally. Ciao -- http://www.dstoecker.de/ (PGP key available) _______________________________________________ Bug-coreutils mailing list Bug-coreutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-coreutils
