Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!cs.utexas.edu!uunet!microsoft!jimad
From: jimad@microsoft.UUCP (Jim ADCOCK)
Newsgroups: comp.lang.c++
Subject: Re: C++ Books: Comments and Recommendations, WANTED
Keywords: book c++
Message-ID: <58238@microsoft.UUCP>
Date: 15 Oct 90 18:56:56 GMT
References: <45531@apple.Apple.COM>
Reply-To: jimad@microsoft.UUCP (Jim ADCOCK)
Distribution: usa
Organization: Microsoft Corp., Redmond WA
Lines: 74

In article <45531@apple.Apple.COM> dks@apple.com (D. K. Smith) writes:
>
>Hi all,
>
>Due to the flood of C++ books out there, I figured I would
>ask for the net opinion on favorable, useful, and useless books 
>on the topic.

The question of books comes up every month or two.  Note that there are
dozens of "C++" books on the market nowadays, that run the full gamot of
quality.  The short short answer remains very simple:

Lippman "A C++ Primer"

still the preferred text for most people's everyday use.  Before posting
to the nets "the facts" about C++, people should consult:

Ellis & Stroustrup "The Annotated C++ Reference Manual"

which is as close to a definitive answer to what the language is that we have,
at least until the standardization committee publishes something.  Total C/C++
neophytes might do well to spend a day or two breezing through the 
programming examples of:

Stevens  "Teach Yourself C++"

For a historical reference to the origins of the language consider:

Stroustrup "The C++ Programming Language"

--but please don't consider it a reliable source for the fine grained details
of the language!  People who post "facts" from the original manual do us
all a disservice.  Consult instead Stroustrup's writings in "ARM" to get
his current statements of what the language is.  Use the original manual:
"The C++ Programming Language" to find out "what it use to be."  

Along with this manual is an excellent text of worked examples:

Hanson "The C++ Answer Book"

Which contains "correct" examples of how to write Int classes, Strings etc.
[many other texts get these "simple" things wrong,  sigh.]  Hanson's text 
concentrates mainly on how to write "simple" useful stand-alone classes.
The other extreme is the excellent recent addition:

Gorlen etc. "Data Abstraction and Object Oriented Programming in C++"

which describes Gorlen's NIH [NIHCL, OOPS] classes, their use, and their
development.  These classes are the "grand scheme" approch ala' Smalltalk
where all classes inherit from a common base class.  

Two other books of general interest, not restricting themselves to C++
[and in fact, their C++ examples have problems] are:

Booth "Object Oriented Design"

covering design methodologies for object oriented programming, and

Khashafian and Abnous "Object Orientation...."

which is an excellent OOP survey book, and also the place to read for issues
of identity, persistence, databases, etc...

There are other good books that closely overlap this list which I do not 
mention.  There are good books targeted to specific audiences that I do not
mention.   And there are lots of "C++" books out there written simply to
capitalize on the rapidly expanding C++ market.  The above list is not
recommended lightly -- I've read almost all C++ books for the general market
that were not transparently bad just from looking at a few pages [And I've
read a lot of books that were transparently bad just because they were
some of the first to market]

Also, note that Jean Chrtistophe publishes a reasonably complete list of
C++ books once a month.