Thanks for your comments. I will pass them on to the ACCU as I think
they need to think about the new interface and other things in general.
I think the new interface is not as good as the old one.
Also your review of the reviews seems reasoned and not just a rant.
The claimed usefulness was in the book reviews. The reviews are, for the
most part worthless or less than worthless.
They are as good, or bad as any other review. Though on the whole
reviews were done by someone actually reading/using the book. Most other
reviews are done by an editor skimming the book or just reading the
cover notes. Although the ACCU editor had to do that for the books that
were not picked up by other reviewers.
BTW I have had evidence that Amazon book reviews can be manipulated.
Also you have no idea who the reviewer is. In one case a lecturer (in
the US) emailed his whole class suggesting that they should put good
reviews (only) on Amazon for his book. The email escaped
You also have no idea who the reviewer is. All the Accu reviews are done
by a names ACCU member. a member listed in the members hand book so we
all know where they live and their contact details. So it is not
anonymous.
A good book getting a bad
review meets my meaning of less than useless.
That is a good book in *YOUR* opinion. The good thing about the ACCU
reviews is (was?) that the appeared in the ACCU magazines before they
went on the web and if people disagreed comments could be made. SO the
book was reviewed by a named person and the review was peer reviewed.
Which is better than most reviews.
I have seen a more than one review challenged, and changed before it
went on the web. The ACCU tended to listen to other users rather than
authors complaints. US authors were particularly bad at complaining at
less than complimentary reviews.
The vast majority of reviews
are by one man, they are not by any stretch of the imagination "peer
reviews". I am convinced that this man wears a green eyeshade, day and
night, and his hobby is counting beans.He has an obsession with
standardization and if there is some minor standards problem in a book the
book will most likely gets a thumbs down, even for one minor infraction.
If you mean who I think you mean that person is not qualified in SW
development and has never worked in Sw development professionally and
his hobby (obsession) is standards so I would have to agree with you
there.
Whilst he has done a lot of the reviews, mainly books no one else wanted
to review to be fair, there are a lot of other reviewers.
The reviews for someone new to programming are unintelligible garbage full
of special jargon no sane "civilian" could possibly make any sense out of.
Some times people like to appear clever by using jargon. It makes them
look like an Expert
I have seen him ignore something Stroustrup, for example, says and criticize
other authors for doing the same damn thing.
Hmmmm..... This is the nature of reviews to some extent, we all have our
heros but I think you have a valid point.
The reviewer feels it is his
bounden duty to say something bad about almost every book.
I must admit some years ago I did go back and re-look at a batch of my
book reviews as I thought I was heading in a downward spiral. However
there are periods where you do seem to get nothing but mediocre books.
The problem is that it is easier and cheaper to produce a book these
days than it was. I think that on average there are more mediocre to
poor out there these days than there were.
Though as you say some like to nit pick to show they are a guru. It can
get obsessive.
A few chosen
authors may escape this crtiicism but it is a rarity. I did not agree with
the reviews given to books I considered the best available at the time. I
would be the first to agree that there area a huge number of bad books on
C++, I have shelves full of them.
As AFAICS The reviewer you are referring has never worked as a
programmer or in any sw engineering environment so he may have different
views to you (or the rest of us) which are more based on the
theoretical, IE standards based, than actual real world software
engineering.
But there *are* good books that didn't
make the cut at ACCU.
The ACCU can, like any other magazine, only publish books given to it
for review. I used to send in any other books that came my way other
than those that were on the ACCU list. Yes, I have been an ACCU book
reviewer off and on for over a decade. Though AFAIK I am not the person
you were referring to previously. (If I am please let me know)
I am reminded of Consumer Reports that had a fetish for leakage current in
appliances. If there were one pico-amp of leakage, the appliance was
rejected out of hand as absolutely unacceptable. Ignore the fact the it had
a great many good properties. Replace "leakage current" with "standards" and
you have a good idea of where I am coming from. Clarity of expression is at
least as important as to whether files that have names such as sdtdio.are
indeed files or might be something other than files. That's an actual
example from memory. Is this really a suitable way to use up five or ten
percent of the space allocated to a two paragraph review? Or is to show off
how incredibly well informed their humble reviewer is?
I have to agree with you here.
The reviews are too brief - a paragraph or two - to be very useful even if
these problems were resolved.
This is historical. In the beginning all reviews were printed in the
ACCU magazines. They had to be short. Then they got put on the web as
well as the magazines. Now I would hope that reviewers do a long web
version and attach a short form version for print.
What little space there is is often wasted
lecturing the author on how the book should have been written. Review the
Gad damned book!! A review is not a place to discuss the philosophy of
writing books.
I agree completely. Some reviewers seem to think they are the star of
the show.
IIRC Richard Heathfield has or had a nice list of books.
DO you have a link?
A person is much better off getting names of some candidate books and
looking at the reviews on Amazon.
I disagree.
They are, indeed, peer reviews. If you are a beginner, your peers are
people that are beginners.
They are reviewed by anonymous people and no one else checks the review.
Also the system is open to abuse. I have documented evidence of at least
one abuse and other evidence of more.
One hundred bad
reviews are better than one bad review. If that isn't the law of large
numbers, it should be. .
Possibly. It depends on the reviewer(s)
Not one review by someone who has been into
computers since the IBM 360 and who has totally forgotten the difficulties
of learning to program. And can, himself, readily spot the difference
between a declaration and a definition, and a parameter and an argument,
and can say "pass by value" without feeling foolish, and on and on and on.
I think the person you referred to previously has no real computing
experience. Other reviewers have.
The idea in the ACCU system was to let beginners select the beginners
books. the problem there is beginners do not always know where they are
being given bad information. It is difficult to get a good review of a
beginners book by some one who can se it as a beginenr but also knows
the problems a beginner may not.
However, no matter how good or bad the reviews the ACCU web site is not
as good as it was.
