In article <983251418.413879.31969.nullmailer / ev.netlab.zetabits.com>, Yukihiro Matsumoto <matz / zetabits.com> wrote: >Hi, > >In message "[ruby-talk:11652] RE: RCR: shortcut for instance variable initialization" > on 01/02/27, Michael Davis <mdavis / sevainc.com> writes: > >|From: Dave Thomas [SMTP:Dave / PragmaticProgrammer.com] >| def initialize(@name, @type, @txt) >| end > >|I like it! > >I don't. I feel it makes the meaning of method arguments ambiguous. >(plus it's ugly.) But it's ok to discuss, since I am easy to change >my mind, you know. I do not understand exactly the nature of the objections. It seems to me to be a fairly innocuous notational shortcut. If I understand correctly the proposal, it is just that when you have a method def f(a,@b,c) ... end it is understood as exactly equivalent to def f(a,dummy,c) @b=dummy ... end unless I missed something? What is ugly? That by analogy someday someone might ask def f(a,$b,c) ... end to be a shortcut for def f(a,dummy,c) $b=dummy ... end I must confess I don't see the problem here either. Please bring light. Best regards, Jean MICHEL