From: transfire@...
Date: 2006-06-14T23:22:07+09:00
Subject: Re: Cut-based AOP


Francis Cianfrocca wrote:
> >>>Well, saying that is sort-of like programming Turbo Pascal in the 80's
> and saying "I have yet to need OOP". AOP is not just a little side
> utility, it's a whole new angle on coding. <<<
>
> I'm looking for ways to put *harder* shells around pieces of well-tested and
> well-designed functionality, not for ways to slice them open even more. The
> article you cite makes a big deal out of "loose-coupling" (the buzzword of
> the decade) as a goal for AOP. I'd rather accomplish that with asynchronous
> messaging.
>
> Austin is right, we don't need AOP.

One can argue the merits (or lack there-of) of AOP all you want. But I
think you and Audtin are being overbarring to tell others they don't
need AOP. That's not your choice. And I don't see Ruby as something so
draconian. There is nothing adverse to the ability to,

  require 'cuts'

if one so chooses.

Implementation unfortuately is a bit tricky, and a pure Ruby
implementation is impossible, but it's not that far off, and I'm sure
if Matz and Peter Vanbroekhoven got together they could iron out a way
for it to be reasonably implemented as an optional .so. Any way you
slice it, Ruby can only benefit from such analystic work.

T.