From: Sean O'Dell <sean@...>
Date: 2003-11-21T14:47:41+09:00
Subject: Re: New Type Checking System Idea

On Thursday 20 November 2003 09:42 pm, Austin Ziegler wrote:
> On Fri, 21 Nov 2003 10:30:06 +0900, Sean O'Dell wrote:
> > On Thursday 20 November 2003 05:24 pm, Austin Ziegler wrote:
> >> On Fri, 21 Nov 2003 09:54:08 +0900, Sean O'Dell wrote:
> >>> It looks like too much happens at run-time, always checking and
> >>> querying for certain methods, etc. Also, the concept of a whole
> >>> interface, rather than querying for a set of required methods, I
> >>> always think is easier to grasp.
> >>
> >> For all intents and purposes, there is no such thing as "compile-
> >> time" in Ruby. Yes, Ruby does go through an AST parser, but
> >> compile-time optimizations won't work in Ruby because any method
> >> can be redefined. At any time, and this invokes compile-time
> >> again.
> >
> > I know, but for my purposes, I think of the point in which the
> > bulk of a main class is loaded as the compile-time. I don't know
> > of another term I can use to refer to this ramping-up period.
> >
> > It would have nice, instead of hearing that I was using the wrong
> > term, you had offered a more correct term.
>
> "Initialization"? "Interpretation"? I don't think that there is a
> good term for it (class definitions are execution of Ruby code), but
> it's rather irrelevant anyway, as the eval'ing of strings or meta
> programming would probably move your checking costs to
> "near-runtime" in any case for large-scale applications.

Okay Ziegler.

	Sean O'Dell