Date: 26 Sep 1996 14:38:51 GMT From: peter@spinner.DIALix.COM (Peter Wemm) To: freebsd-current@freebsd.org Subject: Re: install on {Net,Open}BSD vs install on FreeBSD Message-ID: <52e4hr$fqt$1@haywire.DIALix.COM> References: <199609241530.JAA06226@rover.village.org>
next in thread | previous in thread | raw e-mail | index | archive | help
In article <199609241537.JAA06948@rocky.mt.sri.com>, nate@mt.sri.com (Nate Williams) writes: >> Why is there so much resistance to 10 lines of code already integrated >> and tested in NetBSD and OpenBSD? > > Why is gets() considered to be a 'bad thing'? Because it encourages bad > programming practices when a better solution already exists. > > Why is 'install -d' considered to be a 'bad thing'? Because it > encourage bad installation practices when a better installation method > exists. > > Since the FreeBSD source tree has no need for it (we have a good > solution), and the functionality is available for folks that need it in > our 'provided' sources (GNU-install), then there is no need to 'pollute' > our tree with software that encourages bad practice. Sorry this is late in the argument, but there is one very important hole in the argument... install is *not* specifically designed solely for the purpose of installing FreeBSD build objects into the FreeBSD executable tree. It's a general purpose tool used by a damn lot of other source packages... Just because FreeBSD doesn't need it doesn't mean it's not useful. Saying that "it shouldn't have -d because we have mkdir -p" doesn't hold either because by the same argument, we shouldn't have -C, -o, -g, -m, -s, -f and -c, because we already have cmp, chown, chgrp, chmod, strip, chflags, and both cp and mv. Anyway, that's my $0.02 worth. > Nate -Peter
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?52e4hr$fqt$1>