Re: Book REview

[ok@goanna.cs.rmit.EDU.AU (Richard A. O'Keefe)]

dave@gte.net writes:
>Agreed, but like Feldman said in one of his posts, the first edition
>has been out for several years, and no-one has complained until now.  It would
>seem that his Americanisms do not bother most people.  

Take 500 lines:
	Absence of evidence is not evidence of absence.

(a) How do you know that "no-one has complained until now"?
    For all you or I know to the contrary, _every_ Asian student might
    have complained to his or her lecturer.  All you know is that no-one
    has told _you_ about any complaints.

(b) Nobody *at RMIT* has complained up till now because nobody here has
    used the first edition of the book, or Koffman's Turbo Pascal book
    from which it derives.  Some people don't complain, they just refrain
    from buying the book.

(c) With all of my complaints about Feldman's book, it is head and shoulders
    above *every* C textbook I have seen, and I've seen a lot of them.
    That I spent time raving on about Americanisms instead of gross errors
    in the code and radical misunderstandings about Ada is *good*; in NONE
    of the roughly a score of C textbooks I have looked at did I ever get as
    far as complaining about the index or about cultural bias!

    If lecturers have been using Feldman's books, and not passing on
    such student complaints about cultural bias as they may perhaps have
    had, it could just be an overwhelming relief at being able to use a
    book written by somebody who actually knows what he is talking about
    and can get his basic facts straight that makes them feel it would be
    ungrateful to pass the complaints on.

    Let me quote the October 1992 issue of Scientific American,
    quoting William J. Bennetta, editor of "The Textbook Letter":

	with the exception of some books for advanced placement
	courses, most [US school] textbooks are "wretched junk"
	written by "clowns who know nothing about the subject matter".

    In my view, this opinion could be applied with equal force to most
    of the programming language textbooks I've seen.  (The record is
    held by a small Prolog textbook I once saw in a university bookshop,
    which had an example on nearly every page, and EVERY SINGLE one was
    syntactically illegal.)  For some reason, Ada books seem to be the
    pick of the bunch.

(d) The US has a "culture of dissent"; shared by New Zealand, Australia,
    Canada, and the UK, and some European countries.  In the educational
    context, it seems to be strongest in the US.  The students who are
    worst affected by cultural bias are the very same students raised
    in "cultures of conformity" where it would be *extremely* rude for
    a student to question a teacher or a textbook.  The more affected a
    student is by the issue, the *less* likely he or she is to raise it.

    For example, it is not without signifance that I belong to a church
    which refuses to have an official creed (so I share the "culture of
    dissent") and once memorised the words to "Deck the Halls with
    Boston Charlie" and was planning to use "Rackey Coon Chile" as the
    nickname of my new daughter if she had been a son (so I am pretty
    much Americanised) and had a _great_ time working in California.

    (1) I know which things in the book are "American" and which are
	"Western".  My students don't.  One reason why overseas students
	in Australia might not care to complain is that they don't know
	which things are "host country" matters they would look stupid
	for not knowing, and which things are "foreign" to the host
	country.

    (2) The points which I identified as puzzling are points which
	puzzle someone *friendly towards* US culture and exposed to it
	from childhood up.  If E. D. Hirsch is right, I probably
	understand more of the entries in "Cultural Literacy" than most
	US high school students.  This means that what I have found
	is probably just the tip of the iceberg.  (For what it's worth,
	"GPA" and "sophomore" are not in "Cultural Literacy".)

    (3) I am willing to challenge authority.  I can believe at one and the
	same time the ideas "Feldman is a vastly better teacher than me
	and knows a lot more than I do" and "on a particular point, I can
	be right and Feldman wrong".

    I gave a concrete example of what the "culture of conformity" can do
    to a student when I mentioned an Asian student who was having great
    difficulty with an assignment because he didn't know what the word
    "descendants" meant, and didn't dare to ask his lecturer face to face
    or to raise the issue in the "public" newsgroup for that subject.
    (The lecturer in question is one of the most approachable people in
    the department who has shown great willingness to explain and clarify
    this very assignment.)  Because I had the label "tutor" for this
    subject, because he and I were alone, and *because I asked* him why
    his code had a particular form, he reluctantly explained the problem.


>Nevertheless, I do partially agree with you and Richard:  Either Addison-Wesley 
>should publish special non-American editions or Feldman should add some
>footnotes for the international editions explaining the Americanisms.  However, 
>I do *not* think that he should alter the editions being sold in the U.S. (except,
>of course, the errors should be corrected).

There are two separable issues here.

 -  What ought to be done for Feldman's book in the short term?
    In the short term, reprinting with corrections that don't affect
    pagination is economically feasible; revision is not.

 -  What ought to be done about CS textbooks in general?

    With all due respect, the idea of having separate US/international
    editions is less sensible than it might at first appear.  What are
    just some of the major problems with the idea?

    *  Just because a class is geographically located in the US does not
       mean that the students in it would not benefit from an international
       edition.  I am told that US classes often include a high
       proportion of overseas students or comparatively recent immigrants.

       Don't forget the US citizens raised in Spanish-speaking families.

    *  Having two versions of a book puts up production costs.  Book
       prices in Australia are already high, except for the ones that
       are very high.  The last thing we need is any practice that will
       put the prices up higher.  It's no use having the perfect book
       if the students can't afford it.

    *  Having two versions of a book puts up MAINTENANCE costs.  There
       are two versions to revise for the next edition.  This has got
       to increase the cost, or the likelihood of letting proofreading
       errors slip through, or more likely both.

    *  Nobody has yet presented _evidence_ that the _accidental_
       Americanisms in this book provide any advantage to anyone.
       Will someone have the goodness to explain to me precisely how
       leaving Konrad Zuse and Alan Turing out of a history of milestones
       in computing does anything to help US students?

       Let's face it, if the Americanisms in this particular book were
       deliberate, surely the colour name translation example (program
       3.5) would translate between English and _Spanish_ colour names,
       not English and French.  (By the way, my French teachers would
       have had a fit at the idea that "yellow" and "jaune" could be
       identified.)
	

-- 
Fifty years of programming language research, and we end up with C++ ???
Richard A. O'Keefe; http://www.cs.rmit.edu.au/~ok; RMIT Comp.Sci.