Re: Real OO

[jsa@organon.com (Jon S Anthony)]

In article <68TAah0-3RB@herold.franken.de> jhd@herold.franken.de (Joachim Durchholz) writes:

> > > Well, it's the same with functions and routines. They were introduced
> > > to reuse code (rings a bell, doesn't it?). But nowadays I find myself
> > > using them just for naming a chunk of code.
> > > This is a blatant abuse of the concept of a subroutine; after all, I
> >
> > Why do you say that?  I see no "abuse" of the concept at all.
> > Subprograms stand for a piece of functionality.
> 
> That's not the original intent. Subprograms just happened to be useful  
> for stuff the original inventors didn't consider. I think it's the  
> same for classes; now people say "no, don't abuse classes for modules,  
> that's a far too big gun at a small target".

I really don't think the "original" intent about subprograms is
relevant.  As a concept they originated in "programming" and so if the
meaning has drifted a _tiny_ bit where code duplication is not as
central to the definition, it is not such a big deal.  The idea of
classes and taxonomic models OTOH is at least as old as Aristotle and
I see no particular reason to abuse the notion - especially as class
based OO purports to keep the old semantics of the notion central to
its definition.  I know you can pull a Humpty Dumpty here and make it
be whatever you want, I just think that is very poor practice and adds
to the general decline of communication.


> In language usage, words have a tendency to acquire new concepts if  
> they are halfways appropriate. This has happened with the concept of  
> subroutine, and I expect it to happen with the concept of class.

Actually, they can aquire new "meaning" (or, as is more often the
case, _less_ meaning) simply by being commonly abused.  I just think
such abuse generally has an ill effect on how we communicate and often
suckers neophytes (and even "experts") into confusion.  At best, it 
"just" lowers the effectiveness and precision of language.  A simple
example is the "destruction" of the word "awesome".  A sad case indeed.


> > Yes, I know.  They are (typically) part of LACE (hmmm, if they are
> > actually called clusters, maybe the extra language environment that
> > the implementation is offering has to be called LACE...)  And since
> > it is an add on, it is not as strong a notion and is YALL (yet another
> > little language)...
> 
> Well, I think Bertrand has said what can be said about this. Clusters  
> are a really central concept in BON. Lace is certainly YALL, but BON  
> isn't... and Lace or an equivalent is part of BON.

Doesn't matter.  BON is just YAN and no more integrated than Lace.

/Jon
-- 
Jon Anthony
Organon Motives, Inc.
1 Williston Road, Suite 4
Belmont, MA 02178

617.484.3383
jsa@organon.com