NetNews Usenet Archive 1992 #31/NN_1992_31.iso/spool/comp/os/msdos/programm/11701

Internet Message Format | 1992-12-30 | 6.2 KB

Xref: sparky comp.os.msdos.programmer:11701 comp.lang.c++:18545 comp.lang.c:19031
Path: sparky!uunet!meaddata!ralpht
From: [e-mail] (Ralph W. Trickey)
Newsgroups: comp.os.msdos.programmer,comp.lang.c++,comp.lang.c
Subject: Re: Newbie Wants Advice on C-Programming
Date: 30 Dec 1992 17:29:43 GMT
Organization: Mead Data Central, Dayton OH
Lines: 107
Distribution: world
Message-ID: <[e-mail]>
References: <[e-mail]> <[e-mail]> <[e-mail]> <[e-mail]>
NNTP-Posting-Host: retina.meaddata.com

I would like to add my $.02 to this discussion

In article <[e-mail]>, [e-mail] (Michael Malak) writes:
|> In article <[e-mail]> [e-mail] (Bjarne Stroustrup) writes:
|> >
|> >Consider a few common questions:
|> >
|> >    I don't know C or C++, should I learn C first?
|> >
|> >No. Learn C++ first. The C subset of C++ is easier to learn for C/C++
|> >novices and easier to use than C itself. The reason is that
|> >C++ provides better guarantees than C (stronger type checking).
|> >In addition, C++ provides many minor features, such as the `new'
|> >operator, that are notationally more convenient and less error-prone
|> >than their C alternatives. Thus, if you plan to learn C and C++ (or
|> >just C++) you shouldn't take the detour through C. To use C well,
|> >you need to know tricks and techniques that aren't anywhere near
|> >as important or common in C++ as they are in C. Good C textbooks
|> >tends (reasonably enough) to emphasize the techniques that you
|> >will need for completing major projects in C. Good C++ textbooks,
|> >on the other hand, emphasizes techniques and features that lead
|> >to the use of C++ for data abstraction and object-oriented programming.
|> >Knowing the C++ constructs, their (lower-level) C alternatives are
|> >trivially learned (if necessary).
|> >
|>
|> This is where I respectfully disagree with you.  While programmers
|> undoubtedly do well under your personal tutelage, my experience has been
|> that FORTRAN programmers who teach themselves C commit the cardinal sins
|> against good programming practice that I listed in a previous post.  In
|> my opinion, structured programming, style, and good habits must be
|> taught by a teacher.  And since most structured programming teachers
|> prefer Pascal, I recommended taking a college structured programming
|> Pascal course.

This is where I have to respectfully disagree with both of you.
It is vital to learn structured programming first. The problem is that
there are no courses that teach structured programming, and I don't
think an introductory Pascal course would really cover structured
programming either. I haven't been in college for a while,
but I don't remember any specific course that taught structured
programming, style, and good habits, it was more something that
happened from reading a large number of samples in magazine articles,
and making my own mistakes. I think that you would learn as much from
a modern C course, and it looks better on a resume.

If the only option is to learn from books, I would start with a few
good books that covered ANSI C, not C++. The C++ books that I have
seen make a lot of assumptions about what the reader knows, and throw
too many new ideas at you at once. Your mileage may vary.

Look through the first couple of chapters in the available books, if
they seem understandable, then get the one that you prefer, whether it
is for C or C++. If it is a toss-up, then go for the C++ book.
Remember that you are learning more than one thing, you are learning
both the C syntax and customary usage. Any book will teach you the
syntax, but programming style is more difficult, remember back to the
first programs you wrote, and expect to feel the same way about your
first C or C++ programs next year. You are starting at the bottom of
the curve again.

|>
|> Your points about the superior subset of C in C++ are good, but I don't
|> see anywhere in your book where you caution novices against jumping into
|> class creation right away and just learning the traditional features.
|> Again, one could benefit from having a teacher such as yourself, but the
|> one who is trying to teach himself does not have such guidance and could
|> quickly get mired in the advanced features of C++.  I recommended
|> self-teaching C before C++ because the trick to C++ is when _not_ to
|> create a class, and I've found this mentioned only in articles and never
|> in any textbooks.

It depends, if the person is working on a critical project, he should
definitely use C++ as a better C, and avoid classes except for using
them for data hiding. I learned C++ on my own and made the mistake of
going whole-hog with classes. Luckily, the project wasn't critical. I
learned a lot more from making my own mistakes than I did from reading
books. Especially about the design issues that aren't covered. There
is too much black magic involved in class design on a non-trivial
project.

|>
|> I think the suggestion of a FORTRAN programer jumping into C on his own
|> is a very bad one, and the idea of him getting into C++ is ludicrous.
|>

I agree that it would be a great idea for someone to go through an
intermediate language like Pascal first so they can learn structured
programming, but that won't be practical most of the time.

I think that a programmer would be a lot better off in going to C++ as
a better C instead of C. The problem is that there are better
textbooks out for C than there are for using C++ as a better C. Over
half of a typical C++ book is given to teaching about classes. But he
should use C++ only as a better C until he is comfortable with the
syntax. C++ will protect the programmer from a lot of simple mistakes,
and there are a lot more friendly tools for C++ (like math.h++, and
the streams libraries) that aren't available for C. I agree that he
shouldn't jump right into Object-Oriented Analysis and Design.

--
Ralph Trickey       | [phone]                  |
Mead Data Central   | x4870                           |  Disclaimer
P.O. Box 933        | [e-mail]             |  My opinions are my own
Dayton, Ohio  45401 | ...!uunet!meaddata!ralpht       |