🚀 go-pugleaf

jhd@herold.franken.de (Joachim Durchholz) wrote:

>ell@access4.digex.net wrote 19.03.96 on Re: Beware of "C" Hackers -- A rebuttal to Bertrand Meyer:
>
>> most of the benefits of oo derive from its closeness to objects in the
>> real world.  

>I'd take that with a grain of salt. The real objects may be a useful  
>starting point, but what I see in well-designed libraries are lots of  
>highly abstract objects 

"libraries" is the crucial word. A library of objects is necessarily
somewhat abstract, if you want to see good flexibility and
re-usability.

That isn't to say that objects within a project may not also closely
resemble real-world objects, even if this sacrifices all hope of their
extensibility. The success of "componentware" is an indication of this
approach's value. Componentware may be thought of as objects that have
sacrificed all of their flexibility in favour of immediate ease of
use. The success of tools like VB demonstrates the practical
usefulness of this concept, even if it makes "real" OO designers choke
on their quiche.

I don't see these two interpretations of the object paradigm as
contradictory, even within the same project. The object concept is
broad enough to include both.


If Pascal programmers eat quiche, what _do_ OO designers eat ?  8-)
--
Andy Dingley                          dingbat@codesmth.demon.co.uk                                          

"Cut the second act and the child's throat"
  - Noel Coward, on seeing the young Bonnie Langford on stage