home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!dtix!darwin.sura.net!mips!swrinde!cs.utexas.edu!tamsun.tamu.edu!inetg1!dbsjwb
- From: bielak@arco.com
- Newsgroups: comp.lang.c++
- Subject: Opinions? Convince Mgmt..
- Keywords: Opinions on C++ use
- Message-ID: <1992Jul30.131530.6703@Arco.COM>
- Date: 30 Jul 92 13:15:30 GMT
- Sender: dbsjwb@Edm.Arco.COM (James Bielak)
- Reply-To: bielak@arco.com
- Organization: ARCO Exploration and Production Technology
- Lines: 60
- Originator: dbsjwb@doofus
-
-
- Your opinions and experiences are welcome:
-
- Having had a week-long training in C++, and after reading (much of) Booch,
- Rumbaugh, and Coplien, I am convinced that C++ could be A Good Thing here
- where I work. Using C++ to work with many of the physical and conceptual
- geologic objects I encounter would certainly help application modularity,
- understanding, maintenance, etc.
-
- But there is a problem. Management holds the position "There is no
- standard... We've been burned before using half-baked technology, and
- we don't plan to get burned again" (my apologies if my paraphrasing
- does not exactly express their sentiment).
-
- I learned and wrote K&R C for years before X3J11 arrived, and when it did,
- none of my code "broke". In fact, by the time the standard arrived, I
- was impatient to obtain a "real" compiler which implemented standard C, and
- my style of writing had improved to the point that I doubt I would really
- want to use any of my earlier attempts. If I had waited for ANSI, I would
- still
- be trying to figure out a good use for unions.
-
- Consequently, aside from my personal feelings that I am being left
- in the dust, I spend (too) much of my time trying to translate many of
- my ideas into C code which will implement some of the the utilitarian
- aspects of OO. My List and ListNode, along with the "encapsulation",
- "inheritance" and "objectiveness" with which they are implemented, are
- attempts to create generalized objects which are as uncomplicated as
- possible, yet provide opportunity for reuse and extensibility in the
- immediate future. OTOH, I run into disagreement with another group here
- which claims that the X Windows "Object" class is "the way to go",
- because "X will be here forever".
-
- 1) What arguments can be used for/against allowing development in C++
- today (UNIX workstation shop), in spite of / due to the lack of a
- standard?
-
- 2) If you were to hold the position preventing use of C++, yet wanted
- to encourage conceptualizing objects in C, would you advocate strict
- use of the X Object class, or would you encourage code which utilizes
- OO concepts, is to the point, and gets the job done without the
- X overhead?
-
- Sorry for the long post, and apologies if this topic has been covered before,
- but I feel stuck, and could use your opinions or prior summaries.
-
- Feel free to email any responses, and if there is sufficient interest,
- I will be glad to post a summary.
-
- Thanks,
- jwb... bielak@arco.com
-
-
-
-
-
- --
- bielak@Arco.COM ARCO Exploration and Production Technology
- james.bielak@Edm.Arco.COM 2300 W. Plano Parkway PRC-D3239
- (214) 754-6184 Plano, TX 75075
-