Book Review: Programming C# 3.0 by Jesse Liberty and Donald Xie

re: Book Review: Programming C# 3.0 by Jesse Liberty and Donald Xie

Viktor — Tue, 03 Mar 2009 14:56:35 GMT

I have read previous edition for C# 2.0 and found it inaccurate. I sent my opinion to Jessy. He never responded and did not make errata changes.

re: Book Review: Programming C# 3.0 by Jesse Liberty and Donald Xie

skeet — Fri, 24 Oct 2008 08:13:32 GMT

@Kevdez: Thanks will take a look. Can't say I like the font they've used for headings etc, but I won't let that put me off too much :)

re: Book Review: Programming C# 3.0 by Jesse Liberty and Donald Xie

David Nelson — Wed, 22 Oct 2008 18:56:05 GMT

@Vladimir,

The point is that expressions are not limited to statements. A statement is a syntactic element that has to be able to stand on its own. A program is a sequence of statements. "x = a + b" is a statement and (in C#) an expression (which evaluates to the final value of "x"), while "a + b" is an expression but not a statement. So to define expressions as "statements that evaluate to a value" is to make the definition too narrow.

This goes back to the author's apparent lack of concern for accuracy when using standard terminology. He is apparently using "statement" to mean "any part of the code", but that is not what "statement" means.

re: Book Review: Programming C# 3.0 by Jesse Liberty and Donald Xie

skeet — Tue, 21 Oct 2008 05:27:20 GMT

@Vladimir: I don't think I'd like to try to define an expression in just a few words, really, but possibly something along the lines of "anything in your code which can be evaluated". So in "Console.WriteLine(x + y);" the whole statement (minus the semicolon) is an expression, as is x, as is y, as is x + y.

re: Book Review: Programming C# 3.0 by Jesse Liberty and Donald Xie

Vladimir Kelman — Tue, 21 Oct 2008 05:07:11 GMT

As a programmer who had some limited previous C++ experience and was always interested in OOP concepts, I can confirm, that a combination of "Accelerated C# 2008" (as a main guide) and "C# 3.0 in a Nutshell" (as a reference) is really good. Both books are also extremely interesting.

Jon, you didn't like "statements that evaluate to a value are called expressions". How would you define an expression? In functional programming books they tend to imply that an expression is something which can be evaluated and returns a value when evaluated. Functional languages define a special empty value which is of a special type (like "()" value of type unit in F#).

re: Book Review: Programming C# 3.0 by Jesse Liberty and Donald Xie

kevdez — Tue, 14 Oct 2008 12:25:27 GMT

Hi Jon

Here's another C# book you may care to take a look at and if possible, review.

C# for artists.

Pulp free press.

www.pulpfreepress.com/.../1-932504-07-9.shtml

re: Book Review: Programming C# 3.0 by Jesse Liberty and Donald Xie

skeet — Wed, 08 Oct 2008 13:30:47 GMT

@Rune: I don't yet have much of a recommendation for a "complete newbie" programmer, although I will let slip that I have been talking to a few people about the possibility of writing such a book.

For a programmer who knows Java or C++ but doesn't know any C#, either "C# 3.0 in a Nutshell" or "Accelerated C# 2008" would be a reasonable starting point. I have "Essential C# 3.0" as well, but I haven't read it yet.

One problem is that I think that C# is too big a language to learn from scratch to 3.0 in one book. I suspect it would probably take a year of reading and practising - and I for one don't like the idea of taking a whole year to read one book!

On the other hand, a lot of books which don't cover the language thoroughly don't say what they're leaving out. I hope that if I were ever to write a "beginners" C# book, I would leave appropriate bits out but indicate what's left to investigate, preferably with links to MSDN etc.

(I thoroughly agree with your suggestion that newbies would get nothing out of C# in Depth, btw.)

Jon

re: Book Review: Programming C# 3.0 by Jesse Liberty and Donald Xie

Rune Jacobsen — Wed, 08 Oct 2008 13:11:04 GMT

Hi Jon,

Thanks for another detailed and great review.

Your review and the followup dialogue in the comments between yourself and Paul Batum raises an interesting (to me) question; It's quite clear that someone new to C# (or even worse, new to programming) would probably have a hard time starting with "C# in Depth". After a few years of playing with C# (coming from basic, then assembler, then Pascal, then C and PHP before finally giving in and going OO), I felt that C# in Depth had a lot to offer me; And it probably will the next few times I read it, i.e. it's one of those books that reflect your own perspective so that the more you bring into it, the more you get out of it. However, I can see that a complete language n00b would die before getting through the first few pages;

What would your recommendation be for the book (whether this is a book that already exists, or your opinions on what the book would need to be like) a new C# programmer needs to read before C# in Depth if (s)he is new to

a) programming in general

b) C#

I realize this would not be a review as such, but it would be interesting to see your thoughts about this subject as someone who clearly knows his stuff and has a scary attention to detail.. ;)

Thanks!

re: Book Review: Programming C# 3.0 by Jesse Liberty and Donald Xie

Paul Batum — Mon, 29 Sep 2008 22:48:46 GMT

Thanks for taking the time to provide such a thorough answer Jon. I'm especially looking forward to the next review.

re: Book Review: Programming C# 3.0 by Jesse Liberty and Donald Xie

David Nelson — Mon, 29 Sep 2008 14:42:30 GMT

Calling value types "objects" doesn't really bother me; its really hard to make a definitive case either way. Saying "it's an object when it's boxed" is true only if you equate "object" with "heap reference", which doesn't necessarily have to be true (although if you have a variable of type object it is always a heap reference, which I guess is why we tend to equate the two). However, I can see where it could further confuse the issue for someone who does not already have a firm grasp of the memory model for value and reference types.

The other inaccuracies are inexcusable though. I am in absolute agreement with you: using incorrect terminology because you think that "most real-world programmers" can't tell the difference is exactly why we have so many "real-world programmers" who can't tell the difference! Its like teaching ebonics to grade school students who are already having trouble with English; we should be teaching them the right way, not giving up on them and accepting the wrong way.

re: Book Review: Programming C# 3.0 by Jesse Liberty and Donald Xie

skeet — Mon, 29 Sep 2008 09:19:59 GMT

@Paul: Good question, and one I certainly haven't answered before.

1) I like to provide information for people, particularly where I feel I'm in some ways "better qualified" to provide that information. As I've said before, there are different ways in which a reviewer can be "the right person for the job" - I can't really approach a C# book from the perspective of "How effectively did the book actually teach me" but I can judge its accuracy pretty well. I hope my perspective as an author is also interesting - I probably think more about things like ordering of contents more than others.

2) I like to know my competition - both to learn from it and to differentiate it. As I sort of expected, I haven't found another book which has the same aims as C# in Depth yet - i.e. a really tight language focus. Some other books are "language + framework" focus (e.g. Accelerated C# and C# 3.0 in a Nutshell) - which is absolutely fine, but not what I wanted to write :)

3) I want to improve the experience of people learning C#. By reviewing books I can accomplish that in two ways: I can submit detailed errata to the authors, which should make their next reprint/edition more accurate. I can also hopefully guide potential readers to a book which will suit them well, and give them an idea of what to expect in terms of content, accuracy and style. The better-informed the general C# development community is, the more interesting (and articulate) the discussions on newsgroups and forums such as Stack Overflow is likely to be.

4) I just enjoy doing it :) Now that I'm working in London, I have quite a bit of time on trains, buses and tubes every day - reviewing books is a useful way of spending that time.

5) I hope that by providing detailed book reviews, I'll encourage others to write similarly detailed reviews. I'd really like someone else to give C# in Depth the same kind of treatment I'm giving these books - in particular, I'm sure there are plenty of undiscovered errata, and I'd like to get rid of them :)

Hope that gives some idea of my motivation - it's not a complete answer, but I think it covers most of it.

re: Book Review: Programming C# 3.0 by Jesse Liberty and Donald Xie

Paul Batum — Mon, 29 Sep 2008 08:57:18 GMT

What motivates you to do the reviews? Most bloggers I know will simply write reviews as an incidental "well I just read this so here are my thoughts". But instead you seem to be actively seeking out books that you believe you can review effectively. It seems like an odd choice, as I know I would prefer to read books covering subject matter that I am -unfamiliar- with!

Is it to help differentiate your book (which is FANTASTIC, btw) from the competition? That would make sense.

re: Book Review: Programming C# 3.0 by Jesse Liberty and Donald Xie

skeet — Sun, 28 Sep 2008 21:52:28 GMT

It gets a little bit repetitive, but it's the topic I'm best able to review for accuracy - and of course, the more books I read on a similar topic, the more easily I can compare them.

We'll see how things go though...

re: Book Review: Programming C# 3.0 by Jesse Liberty and Donald Xie

Paul Batum — Sun, 28 Sep 2008 21:31:27 GMT

Really enjoying these reviews Jon. I must admit though, I'm surprised that you are prepared to review so much material that is all heavily C# related - does it get at all tiresome?