Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!tut.cis.ohio-state.edu!cwjcc!alpha!himanshu
From: himanshu@alpha.CES.CWRU.Edu (Himanshu Rawell)
Newsgroups: comp.sw.components
Subject: Re: Ted Dunning's flamage (dousing)
Message-ID: <298@cwjcc.CWRU.Edu>
Date: 28 May 89 21:19:27 GMT
References: <698@oliver.analogy.UUCP> <8630001@hplsla.HP.COM>
Sender: news@cwjcc.CWRU.Edu
Reply-To: himanshu@alpha.UUCP (Himanshu Rawell)
Organization: CWRU Dept of Computer Engineering and Science, Cleveland, OH
Lines: 35

In article <8630001@hplsla.HP.COM> jima@hplsla.HP.COM (Jim Adcock) writes:
>> ..in the near future. How about ideas for C? 
>> And I don't mean "switch to C++". 
>


>Why not switch to C++ ???

********* WHY NOT C? ************

Why put cart before the horse? I have implemented my entire
Master's thesis in C (35,000 lines of C code) mostly by
restructuring and modifying the existing C code. Difficult, but a
sensible and productive thing to do (Reverse Engineering). Reusability
in action (C++ => C, is reinventing the wheel).

By the way, my thesis is "A Catalog of Six Families of Reusable Software
Components". There are six components in each family. Some examples are
-   Buffer mechanism in EMACS-like text editors
-   Screen Update mechanism in EMACS-like editors.


>
>One could still write reuseable packages in C++,

Hey, why write them? Just modify the existing ones, to make them
have encapsulated interfaces. We are talking reusability. Again
one can write encapsulated components (object-oriented lingo!)
in C, by using disciplined programming. This is to avoid messing up
all the existing software and still have the reusable components around.

>and give them a C compatible calling interface
>using the extern "C" construct of C++.
>
>???