On Sep 7, 2010, at 5:21 PM, NARUSE, Yui wrote:

> 2010/9/8 Joshua Ballanco <jballanc / gmail.com>:
>> I've been itching to start discussing/debating/learning about =
features for Ruby 2.0 for a while now.
>> Some of the recent threads on ruby-core have just reemphasized the =
point. So, I was curious
>> what the process for 2.0 might be? I notice that there is not an =
entry on Redmine for 2.0 yet.
>> Will there eventually be a wiki? What is the preferred method for new =
feature suggestions?
>> Should we open tickets? Create a wiki page? Start a thread on the =
mailing list? All of the above?
>=20
> What is Ruby 2.0 is still ambiguous.
> I think it is depend on compatibility; if some feature is incompatible =
with 1.9,
> the new version which include them will be 2.0.

So, for example, a few things I've wanted for a long time:

-- "def" returns a lambda instead of nil
-- adding parens "()" to a lambda has the effect of calling it (i.e. =
my_lambda.() --> my_lambda() )
-- full access to subclass from metaclasses (i.e. allow prototype-style =
inheritance)
-- true namespacing (i.e. adding a layer to const lookup without adding =
a class/metaclass)
-- named parameters

While these aren't necessarily incompatible (well, except maybe the =
parens for lambda), they are language-level changes. I imagine they =
should be considered carefully and deliberately and not necessarily in =
the context of a point update. I hadn't ever really made any fuss about =
these issues because I felt that they would be more appropriately =
discussed in the scope of Ruby 2.0, and I still believe that. Also, I'm =
willing to be patient (no need to get into a war of words now), I'd just =
like know how/when it would be best to raise these issues for =
consideration.

Cheers,

Josh=