On Monday 24 November 2003 09:19 am, Tanaka Akira wrote:
> In article <200311230648.41003.transami / runbox.com>,
>
>   "T. Onoma" <transami / runbox.com> writes:
> > I don't see how this is a big deal. When I first read this on Elliots
> > post I thought "so?" You can mispell things anywhere and its going to
> > break stuff. All you're pointing out is that no definite error is going
> > to pop out and say "hey dodo, you made a typo!".  Well, guess what? That
> > kind of bug happens all the time, especially in a dynamic language like
> > Ruby. So I think that's a very very small point, and thus doesn't
> > invalidate less code argument. Besides what do you do if no block is
> > assigned?...more code.
>
> I see.  You don't care typos.  But I care typos.

I would not say I don't care. I just mean that for me it is not a "shop 
stopper" of other benefits.

Also, it is an interesting exploration of a new idea.

> I decided the interface: symbol keyed hash with explicit option name
> validation.  (The another idea in [ruby-core:1709])
>
> It is a simplest way to do it.  I think other proposal's benefits
> doesn't justify their code bloat and maintainability loss by their
> complexity.
>
> In future, I may reconsider the proposals if the simplest way cannot
> fulfill requirements, but I don't take them now.

Great! I think that will work fine.

Thanks!
-t