From: gabriele renzi <rff_rff@...>
Date: 2004-10-10T20:14:40+09:00
Subject: Re: ANN: Free-form-operators patch

Nikolai Weibull ha scritto:

> * gabriele renzi <rff_rff@remove-yahoo.it> [Oct 10, 2004 03:00]:
> 
>>>+= is given by +, so that's no problem.
> 
> 
>>actually, I'd like to override += (just think of Array#+= done via in
>>place addition or with sum and reassign), just to say :)
> 
> 
> Well, there you go.  Your whole reasoning just went down the drain.  If
> this would be allowed, how would anyone keep track of how += for
> arrays differs from += for, say, integers.  

Sorry but you misunderstood my point completely.
I don't want to keep track, I expect people to do the right thing just 
like they do now (i.e. not redefining Fixnum#<=> to be the opposite of 
Bignum#<=>).

I expect that if people would override += they would do this sanely.

> It's more confusing to alter
> the meaning of some "well understood" operator than to add new ones with
> new semantics,


I am not against the concept of new operators, I said I like the idea.
And I do not expect people to be idiot, where they would override 
important operators they would probably do this for some good reason 
(i.e. Criteria-like things would benefit from overriding !=).

What I am afraid of is the interaction beetween different users's 
language extensions, in a space where name collision is more easy than 
in the method's one.

But there you go, we both exposed our reason let matz decide :)