On Mon, Jun 2, 2008 at 8:03 PM, David Masover <ninja / slaphack.com> wrote:
> Interesting that this works -- but it doesn't disprove my point. Humans are a
> category of mammals -- that doesn't imply that all mammals are humans.

Humans are a category of animal with certain characteristics which set
them apart form other animals.  All animals which have those
characteristics are human.  You stated the theory that the NilClass
class encompasses all objects that respond positively to #nil?.  Which
is easily disproved by defining a non-NilClass which responds
positively to nil.  MyNil is in the *category* of objects which
respond to #nil? in the affirmative, but not in the *class* NilClass.
Thus, NilClass != the category of objects which respond to #nil?

NilClass objects may be a *subset* of the #nil? => true category, but
NilClass itself does not define a category other than "objects which
are instances of NilClass".

-- 
Avdi

Home: http://avdi.org
Developer Blog: http://avdi.org/devblog/
Twitter: http://twitter.com/avdi
Journal: http://avdi.livejournal.com