From: Jacob Fugal <lukfugl@...>
Date: 2006-08-02T06:10:49+09:00
Subject: Re: I thought this was the one that worked?

On 8/1/06, Chad Perrin <perrin@apotheon.com> wrote:
> On Wed, Aug 02, 2006 at 05:33:50AM +0900, Jacob Fugal wrote:
> > On 8/1/06, Chad Perrin <perrin@apotheon.com> wrote:
> > >That's an impressive bit of gymnastics.
> >
> > Gymnastics? What exactly did I twist?
>
> My statement has been divorced of context by separating it from the rest
> of my brief paragraph.  That context is necessary to understanding of
> the statement, and answers your question.  Please don't do that.

I apologize. But in my defense, the rest of the paragraph (which
immediately follows) doesn't explain what potion of what I did was
gymnastics, nor how. That's why I separated it.

> > >I don't buy it as satisfying
> > >the requirements of a closure, though, at this point.  Self is just
> > >something within the current scope, and we're back to an absurdly broad
> > >redefinition of "closure".
> >
> > What current scope are you referring to here? If you mean the scope
> > that was current when the block was created, then why does that not
> > make it a closure? Why is self different than some other variable
> > named b? If you don't mean the scope that was current when the block
> > was created, then you're mistaken, because it *is* the self that was
> > current when the block was created that is used. This is what others
> > were trying to demonstrate in the examples you dismissed as having
> > irrelevant variables.
>
> When you say "the scope that was current when the block was created",
> are you referring to the scope that encloses the code that would become
> a closure, or are you referring to the scope of the would-be closure
> itself?

The scope that encloses the lambda, which was introduced by the def
keyword immediately preceding the identifier "bar". Specific enough?

> > >> There is no such thing as a "function" in ruby; they are all methods.
> > >
> > >Methods are "functions" in the technical, broadly interpreted computer
> > >science definition of the term.  They're just a special case of
> > >"function" that requires more specific conditions to qualify as
> > >"methods".  A method is a function: a function is not necessarily a
> > >method.
> >
> > Ok, we're using different definitions of function. That's why I put
> > function in quotes when I said it, but I can see you want to be
> > pedantic. By your definition, a method is indeed a function.
>
> I'm trying to be precise.  It's lack of precision that got us into this
> mess in the first place.
>
> > But I counter your last statement: in Ruby, all functions are
> > necessarily methods, because you can't call a method/function without
> > a receiver. You can write it syntactically without a receiver, but in
> > that case a receiver of self is still implicit. In this sense, Ruby
> > has no "functions", only methods.
>
> That doesn't counter my last statement at all, since I didn't say "a
> function is not necessarily a method IN RUBY".

Well, my original statement did include the phrase "in Ruby", so I
thought it appropriate to try and interpret your statements within
that context, since we are discussing functions and closures IN RUBY.

> I was referring to
> functions in the generic.  To properly sort this out, we need to refer
> to something approaching a formal definition of a closure (which
> includes more than just Ruby, and thus necessitates using terms in a
> manner that apply to more than just Ruby), or we need to simply state
> that the term "closure" has a different meaning in Ruby than elsewhere.
> Take your pick.

Unfortunately, I don't have a reference text on me, nor can I find a
source online that you might consider authoritative, so we'll just
have to leave the formal definitions for later. But the version of
closures I learned and implemented in the University were those which,
as Ara mentioned, contain some sort of pointer to the environment in
which they were created so that they can perform name resolution using
that environment. All procs/lambdas in Ruby satisfy that condition.
The discussion of functions and methods in Ruby, imprecise as it may
be, is incidental to that, so I felt no need to be that formal or
specific about what a "function" is.

Jacob Fugal