From: "Hal E. Fulton" <hal9000@...>
Date: 2003-03-29T05:21:46+09:00
Subject: Re: Array question

----- Original Message ----- 
From: "Michael Campbell" <michael_s_campbell@yahoo.com>
To: "ruby-talk ML" <ruby-talk@ruby-lang.org>
Sent: Friday, March 28, 2003 11:58 AM
Subject: Re: Array question


> --- Yukihiro Matsumoto <matz@ruby-lang.org> wrote:
> > Hi,
> > 
> > In message "Re: Array question"
> >     on 03/03/29, Michael Campbell <michael_s_campbell@yahoo.com>
> > writes:
> > 
> > |out of curiosity, why would a block to #join be awkward?  It's
> > wholly
> > |additive to the library, it doesn't break any existing code, and
> > |anything you write could just as easily not use it.
> > 
> > Somehow, I feel feeping creaturism here.
> 
> I'm not totally disagreeing, and I'm not sure that we should just add
> things because we can.  *BUT*, my question here was why something
> like that would feel awkward to the original poster?  It won't change
> a single thing in his existing code, nor force him to do anything
> different in the future.  It is, for all intents and purposes,
> invisible.

My opinion only.

It's an obscure feature, very rarely needed, 
and it is not intuitive that such a feature
would even be there.

There are many methods that don't take blocks.
They could *all* be revised so that they in
effect modified the receiver or did a map in
some way before doing their real job.

I wish there were a way to measure "usefulness"
of a feature and "complexity" of the interpreter
in a totally objective way.

Then I would say that the increase in usefulness
has to equal or exceed the increase in complexity
in order for a change to be viable.

Cheers,
Hal