The Herbert Schildt wikipedia article has been today nominated for
deletion, and not by me. I have however said that owing to its poor
sources it is in serious violation of wikipedia's own policies.
Here is the text of my case for Delete in the rather likely event my
text is itself deleted:
"Thou shalt not bear false witness against thy neighbor" - Exodus, KJV
The Seebach and Feather posts are clearly polemical. Basically, both
went through a Schildt book in one pass trying to find as many
"errors" as possible. Both make their own errors, including the false
claim that "void main() is not standard C" when according to the C99
standard it is indeed, just not "hosted": it is freestanding standard
Errors in a computer book's code examples are a somewhat serious
matter, rather like errors in commercial software, and to-date no
practical method has been found for avoiding either. However, McGraw
Hill, like any software or computer book publisher, indemnifies itself
through a warranty disclaimer concerning errors. This means that the
programmer-reader, to get the benefit out of what are intended, in
Schildt, to be representative code snippets, needs to exercise
caution, and learn, while typing those code snippets into a particular
implementation of C.
All computer books contain such errors as a byproduct of the author's
human limitations, the production process in which live code becomes
dead PDF, and the stability of the particular programming language
being discussed. It's easier to err in the case of C, which has never
been responsibly standardized and in which aliasing creates
instability, to make "errors".
As in the case of Kathy Sierra, programmers who are rather aliterate
(as is evident from Seebach's and Feather's strange use of "clear"
when they call Schildt "clear") tend to be confused by a breezy style
and prefer manuals which the mere mortal cannot understand. Seebach,
who led the charge against Schildt, confesses to having a radically
different learning style in which he is easily confused if something
is expressed in a non-literal way.
But Herb's intended audience does not learn in this way. They
understand, and at times love, the goofy professor who gets a proof
wrong, and uses his own mistakes to teach something new. They
appreciate a chance to try a code snippet, find that it works wrong,
and fix it.
Here is wikipedia's own policy:
We must get the article right. Be very firm about the use of high
quality sources. All quotations and any material challenged or likely
to be challenged must be attributed to a reliable, published source
using an inline citation. Contentious material about living persons
that is unsourced or poorly sourced=97whether the material is negative,
positive, neutral, or just questionable=97should be removed immediately
and without waiting for discussion.
Not one of the three sources are of sufficient neutrality to be
reliable. Seebach starts by accusing Schildt of having written a book
with hundreds of bugs (but then presents only twenty in the previous
edition of "C: the Complete Nonsense", and only a few more in the
next). The bugs turn out for the most part to be artifacts of the
instability of C and Microsoft/Linux differences.
Feather's document as a copy-cat, drive by shooting emulates Seebach:
it's a claim that "this author is bad" with less than fifty examples
of why he's bad, most of which are trivia and violation of Linux
The ONLINE copy of the Summit FAQs does not even reference Schildt;
instead it snarkily refers to a fictional book, with the insider joke
that this means "Schildt".
The sourcing of the Schildt article is extraordinarily in violation of
Some of the posters below say that "Schildt's friends can post
favorable reviews". This however, reminds me of the kangaroo court I
was subjected to in 2006 when I was bullied by wikipedia editor
amerindianarts; out of the blue one finds one is on Trial. In real
law, bringing a charge is considered a serious matter, for a grand
jury. Here, anyone can ruin anyone's life by bringing a charge to
which the person has to respond. There's plenty of favorable
information on Schildt, of course, starting with his sales figures and
his adoption as a textbook. But he should not have to stand
trial...unless all computer authors must stand trial, such as myself,
or Peter Seebach.
I can see below that Schildt's enemies would like this review to be a
plebiscite. This is however to be ignorant of the law of small
numbers. The people who vote to "keep" are too small in number to
constitute a plebiscite, and too invested in a pro-Linux outcome to
constitute a jury.
Finally, this strange matter that a monstrum horrendum, a sock
puppeteer, and a ruffian like myself should be also a white knight,
and as such, as an Emile Zola defending no friend of his, a Dreyfus-
Schildt, should act in such a disinterested way. It is because my own
defense against my bullying on wikipedia is futile because snot-nosed
convenience store clerks don't like my prose style, having been ill-
equipped to read above a low upper bound of complexity. Whilst still
being adequately prolix relative to the issues at hand, up to and
including the Fascism of Wikipedia, I find it more effective to undo
the damage done to a hard working computer author.
As a hard working computer author, family member, and member of his
community, Herbert Schildt's right to privacy, guaranteed to him by
the Ninth Amendment of the United States Constitution, have been for
too long violated by this article. You don't have an article about me,
although I'm a computer author. You don't have an article about Dan
Appleman, who's written extensively on computers and is a real nice
guy. And this is as it should be. Computer authors are for the most
part employees of computer publishers who hew closely, as did Schildt,
to a marketing plan. They are not Zolas, able to publish their own
views at will; they are more like Captain Dreyfus, honorable men and
women who try to do their best.
Ecclesiastes says "let us now praise famous men, and their children
after them". It goes on to say that we must honor men who are
invisible, who raise families and work hard at their jobs. If they
find they can actually write more than the disorganized hate mail of a
Seebach or a Feather, they discover, as I discovered, that they can
make a little extra cash writing books about their trade, perhaps to
send their children to school.
We honor them by leaving them alone, and not dragging their name in
the mud. You bear false witness against obscure men when on the basis
of the superstitions shibboleths of an unstable programming language,
you make their father's name, their sons' name, their wive's adopted
name, into a foul word, such as "Bullschildt".
Take this article down.
Edward G. Nilges