From: JamesBritt <james@...>
Date: 2002-10-13T02:46:34+09:00
Subject: Design Philosophies (was Re: Different Behavior of Class Variables w.r.t ||=)



> -----Original Message-----
> From: Gavin Sinclair [mailto:gsinclair@soyabean.com.au]
> Sent: Friday, October 11, 2002 8:07 PM
> To: ruby-talk ML
> Subject: Re: Different Behavior of Class Variables w.r.t ||=
>
>
>
> From: " JamesBritt" <james@jamesbritt.com>
>
>
> > > I think Ruby should have them on by default, and use a
> > > command-line switch to suppress them.
> >
> > Where do you put the switch if you've set a file as executable
> and run it
> > without specifying the interpreter?
>
> Fair enough.  In Unix, of course, that's a non-issue.  But
> there's a way around
> that as well...
>
> > Why not just set your RUBYOPT value to include warnings?
>
> ...why not set RUBYOPT to suppress them.
>
> My point is not for myself; I *always* use -w.  *Always* :)  My
> point is for
> other people.


This is an interesting question, analogous to the forced use of white space
in Python or the absence of pointers in Java.  The philosophy seems to be
that it's better to force certain behavior for a developer's own good.

If you haven't been using the -w switch all along, Setting RUBYOPT=-w  is
akin to pulling back your couch after not cleaning for several months.  You
find all sorts of icky things.  And, as with your couch, you have to wonder
how much of the ick is actually desirable.

When you see a list of warnings for a given script, how often do you say,
"Yeah, I know, I *meant* that"? Do people here have programmatic idioms
that, while producing warnings, are justified?

I'm not fond of restricting programmer choice; there should be enough rope
to shoot oneself in the foot.  But so long as warnings can easily be turned
off, what are the arguments for not forcing people to look under the couch
all the time?  What do you gain by avoiding warnings by default?


James