Message-ID: <39EB6A28.C4EF19D9@eton.powernet.co.uk> Date: Mon, 16 Oct 2000 21:50:48 +0100 From: Richard Heathfield Organization: Eton Computer Systems Ltd X-Mailer: Mozilla 4.6 [en-gb]C-CCK-MCD NetscapeOnline.co.uk (WinNT; I) X-Accept-Language: en-GB,en MIME-Version: 1.0 Newsgroups: alt.comp.lang.learn.c-c++,comp.os.msdos.djgpp,comp.programming Subject: Re: Undertaking a programming journey References: <8scg36$gsm$1 AT nnrp1 DOT deja DOT com> <39E9CF07 DOT 785C0C0F AT eton DOT powernet DOT co DOT uk> <8scls9$kth$1 AT nnrp1 DOT deja DOT com> <39E9FAD5 DOT DE1FDAE4 AT eton DOT powernet DOT co DOT uk> <8sdrub$h7u$1 AT nnrp1 DOT deja DOT com> <39EAA40B DOT 31B0CA89 AT eton DOT powernet DOT co DOT uk> <8seoli$65v$1 AT nnrp1 DOT deja DOT com> <39EAF73E DOT ECA52E1A AT antlimited DOT com> <8sfbu7$n06$1 AT nnrp1 DOT deja DOT com> <39EB4271 DOT 85CE6874 AT antlimited DOT com> <8sfhr4$si2$1 AT nnrp1 DOT deja DOT com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit NNTP-Posting-Host: 195.60.5.107 X-Trace: 16 Oct 2000 21:55:03 +0100, 195.60.5.107 Lines: 113 To: djgpp AT delorie DOT com DJ-Gateway: from newsgroup comp.os.msdos.djgpp Reply-To: djgpp AT delorie DOT com ChuckEasttom wrote: > > In article <39EB4271 DOT 85CE6874 AT antlimited DOT com>, > Richard Heathfield wrote: > > [Hey, Chuck, do you think we should start selling tickets? :-) ] > > > > Absolutely!:) You print and distribute them, and I'll collect the money. ;-) > > [Herbert Schildt] is a persuasive author; this should be a good thing but, > > unfortunately, in his case it is not. He covers difficult topics with > > aplomb - perhaps this is a consequence of his Masters degree. What a > > pity, then, that I cannot trust his code. > > > I am not saying his code or his books are perfect. I am saying, > however, that your statement "I cannot trust his code" is simply too > strong. I don't think so. Now, I am quite sure that an experienced C programmer could take the ideas Schildt presents, and make workable, or even excellent, programs out of them. What is at issue here is that the flawed techniques Schildt demonstrates are aimed at /newcomers/ to C, who are unable to distinguish between what is correct and what is not. All too often, they happen to be using the very platform he assumes they are using, with the result that the code often appears to "work", even though it's flawed. Not all environments are as forgiving as Wintel, so newcomers to C learn techniques which suddenly stop working when they switch platforms, and they don't understand why. > > > > > > 3. And president of a consulting firm. > > > > Yes, a lot of us are in that position. > > Congratulations. Not really - I am the only consultant in the firm! Now, Schildt may be the president of a firm employing 200 people. Or he might be a one-man band, like me. I suspect the latter, but have no particular evidence for that suspicion. > Oh I will undoubtably be scouring your book for errors:) Good. That's what peer review is /for/. > > But my point is this: Your book has some errors, Schildts book has some > errors... I have a little epublished VB book that I am currently > revising because of...yep errors. But your statements seem to indicate > that if his books have any errors, then they cannot be trusted. No, I've already pointed out in this thread that errors are inevitable. What marks Schildt out as special is: 1) Simplicity. Many of his mistakes are howlers that one would expect from a newbie, but not from a best-selling author. 2) Platform-dependence. Many of his mistakes happen to work on the platform he uses, and assumes his readers use - his attitude seems to be "if it works on my compiler, it must be right". In a book on Windows programming, this might be acceptable. In a book on ANSI C programming, it is not. 3) Intransigence. When people have, in the past, pointed out his mistakes, he has denied that they are mistakes at all, despite firm proof. (I'm thinking particularly of when the ANSI committee tried to point out to Mr Schildt that main returns int. It has taken him many years to come round to the idea, and I suspect he isn't totally convinced even to this day.) 4) Abdication of responsibility. If Mr Schildt publishes an errata list, I certainly haven't found it. If I'm wrong about that, I hope someone will point out the fact, for I'd be glad to be proved wrong on this count. Everyone makes mistakes. But we expect that authors will make advanced mistakes, not simple mistakes. We expect that authors of books on ANSI C will take care to ensure that their programs work on, at the very least, two reasonably different platforms. We expect authors to be big enough to accept when they're wrong. And we expect authors to provide their readership with a list of errata in published works. At least, those are my expectations. Perhaps I'm just weird. > I guess that is my real compaint. I would think that professional > people, most of whom have a formal education, could express their > criticisms in a very civillized manner. Yes, you'd think so, wouldn't you? And, do you know what? To a large extent, this *is* in fact the case. People are considerably more polite on Usenet than I actually expected would be the case. There is still rudeness, of course, and I'm quite sure I'm guilty of it myself. But it's naive to expect people who answer questions regularly to be unfailingly charming and polite to those who don't bother to follow the normal conventions of Usenet in general and, more particularly, the customs and norms of the newsgroups to which they post. > Well perhaps that is my problem..I just don't have the stomach for > usenet. I hope that this very thread has shown that it is possible for people to conduct a civilised discussion on Usenet, even when they are, to some extent at least, in confrontation with each other. -- Richard Heathfield "Usenet is a strange place." - Dennis M Ritchie, 29 July 1999. C FAQ: http://www.eskimo.com/~scs/C-faq/top.html 66 K&R Answers: http://users.powernet.co.uk/eton/kandr2/index.html (31 to go)