Date: Sat, 2 Apr 2011 10:19:43 +0000 From: Alexey Dokuchaev <danfe@FreeBSD.org> To: Alexander <freebsd@nagilum.org> Cc: svn-src-stable@freebsd.org, Xin LI <delphij@gmail.com> Subject: Re: svn commit: r220250 - in stable/8: bin/test tools/regression/bin/test Message-ID: <20110402101943.GA81206@FreeBSD.org> In-Reply-To: <20110402115336.37394ztrsha7r58o@cakebox.homeunix.net> References: <201104011823.p31INiHv055195@svn.freebsd.org> <20110402015552.GA32901@FreeBSD.org> <BANLkTinubz%2B00HneNukiuZTeOMXfu0=JtQ@mail.gmail.com> <20110402115336.37394ztrsha7r58o@cakebox.homeunix.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, Apr 02, 2011 at 11:53:36AM +0200, Alexander wrote: > The downside is that one may end up writing scripts on FreeBSD which > will unexpectedly break on other Unixes. Exactly. I also fail to understand why would we want to strive for compatibility which broken shells (bash) by introducing controversial features that are not part of any standard and deliberately undocumented. This is what usually happens when deprecating stuff, no? Introducing new functionality in "stealth" mode like this makes little sense, and I believe does more harm than good. I recall that it was discussed here on svn-* lists after in was committed to HEAD, and opinions were mostly negative IMSMR. I am a bit surprised it was not backed out (not that it is my call). ./danfe
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20110402101943.GA81206>