This is a multi-part message in MIME format.
-----214afe3d3b94532dfc29ee94c3abb74e
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit



Original Message:
-----------------------------------------------

>>On Sun, 24 Aug 2003, Chad Fowler wrote:
>>
>>> Matju, is there any reason to do this other than idealism?
>>
>>Chad, wasn't the module/class distinction a similarly "idealistic" move in
>>the first place? I mean it seems to be on the same level of thought than
>>what I propose. Then why is the so-called "idealism" inappropriate when it
>>comes from me?
>>

I didn't say that idealism isn't OK.  But, it's a lot harder to change something due to idealism than it is to create something idealistically.  I fully understand the desire to unify design concepts in the language, but I haven't heard anything that makes it seem worth it at this stage in Ruby's life.

>>> We've beaten this thing to death in terms of why *not* to do it,
>>
>>I can't share this opinion. A lot may have been said, but I suspect that
>>not so much has been understood, and certainly that few people have
>>participated in it. You seem to have a different standard on what "beating
>>to death" may mean.
>>

Obviously. :)

>>> But, why *should* this change?
>>
>>I've presented this already. [ruby-core:1448]. The actual advantages may
>>get more precise as the details get more precise.
>>

I read ruby-core:1448 both when it was originally sent and today.  As I said, I can understand point #1 (unification of concepts).  Points 2-4 are subjective, and I don't know many (or really anyone else) who would agree with them.

>>> What is the real practical benefit?
>>
>>You mean that the reasons I've presented yet are not practical enough for
>>you. I'm sorry I can't help you for now, especially as I don't completely
>>understand the current situation yet, even after "beating this thing to
>>death" as you call it.
>>
>>> What could we do that we can't do now?
>>
>>Not all changes worth doing fall into that category. Else chances are that
>>I would have sticked with another language and never would have bothered
>>with Ruby.
>>

Certainly not all choices are based on tangible benefit.  But the choice to change a language that thousands of people are using should hopefully be backed by some kind of tangible benefit.  If it's true that the current choices were made based on idealism and yours are as well, they cancel each other out.  Not to mention the fact that the initial choices were made by Matz. :)

Still interested in hearing a tangible reason to make the change (maybe an example of where real MI would be helpful?),
Chad




-----214afe3d3b94532dfc29ee94c3abb74e--