In article <4870beef$0$18153$4fafbaef@reader3.news.tin.it>, nembo@kid
says...
Quote:
According to accu.org, Deitel's "How to program" is not reccomended.
>
I don't agree. I have the 3rd ed. and I find it ok.
>
What do you think about this book?
|
The 3rd edition is pretty old. I was at the bookstore today, however,
and took a look at the current (6th) edition of the same book. I
certainly didn't try to read the whole thing (well over 1000 pages) but
glancing through bits and pieces for 15 or 20 minutes did not leave a
particularly good impression. While better (for example) than any of the
Herbert Schildt books I've looked at, what I looked at wasn't exactly
great either. In almost every H.S. book I've glanced through, you could
open to a more or less random page and find at least one thing that was
clearly _wrong_. In this book I wouldn't say I saw anything that was
clearly wrong but a fair amount of what I saw wasn't particularly
_right_ either. The precepts of the examples weren't particularly
realistic, and the designs of the example classes were mediocre (at
least the ones I glanced at, which included a string class, a date class
and a hierarchy for employees, including derived classes for salaried
and commissioned employees).
While I wouldn't really recommend _against_ it, I think there are
substantially better books such as Francis Glassborrow's books for rank
beginners, and Accelerated C++ for anybody with prior experience. If
you're looking for a good book, there are better ways to spend your
money -- but if you already have it, you could do worse than reading it
and doing the exercises.
--
Later,
Jerry.
The universe is a figment of its own imagination.
Why Deitel's book should be bad? 
