From: Ilias Lazaridis <ilias@...>
Date: 2005-05-12T10:15:27+09:00
Subject: Re: [RFC] traits-0.1.0

Ara.T.Howard@noaa.gov wrote:
> On Wed, 11 May 2005, Ilias Lazaridis wrote:
> 
>>> - trait is short
>>
>> and used within C++
> 
> hmm.
> 
>>> - traits is a synonym for attribute (attr, attr_accessor, etc)
>>
>> no need to use a synonym.
> 
> i mean it's meaning is the similar to 'attr :a == trait :a'.

I mean: no mean to use (and search) a synonym

>> [...]
>>
>>> random thoughts for attributes/against traits
>>>
>>> - no confusion with other traits
>>
>> yes
> 
> but are the other traits being used?  also these seem to be called roles in
> some langs too...  this is a valid concern though.

again: no need to research.

>>> - the alias 'has' can make 'attribute' plenty short (the current 
>>> version of traits uses this alias too btw.)
>>
>> yes
>>
>>> one other thing, the impl of traits was suprisingly hard.  the tricky
>>
>> [...]
>>
>> The code fragments (usage) I've saw are not thus attractive.
>>
>> Code should looks simpler, more 'natural'.
> 
> 
> what do you think could be more natural than
> 
>   class C
>     has 'a' => 42
>   end

class C
   a = 42
end

class C
   var a = 42
end

>   c = C::new
>   p c.a       #=> 42
> 
> or
> 
>   class AbstractWidget
[...]

>   class SmallishBlueWidget
[...]

I dislike all this, but cannot clarify why.

concept, ruby, flexibility

> it's the best i could manage.  come up with something more natural and i'll
> see what i can do to implement.
> 
>> This is an important project, but should be setup collaborative.
> 
> i'm working 70+ hours per week already - but if someone wants to take 
> over and do the initial setup work it's fine with me.

I think you should postpone this project.

>> So, if you setup a ruby-forge project, people can contribute.
> 
> this is welcome of course - but see above.
> 
>> call it "attributes" or "attrib" or "has" - but not "traits".
> 
> +1 for attributes then.

ok

..

-- 
http://lazaridis.com