Writing Programming Books

"persius" <mark.bower@persius.freeserve.co.uk> wrote:

>I'd like some advice on writing books for developers. If anyone has any
>experience of working as a technical author perhaps you'd help me out with
a
>few guidelines.

As an author of an advanced level C++ book ( "The ANSI/ISO C++
Professional Programmer's Handbook",
http://www.amazon.com/exec/obidos/ASIN/0789720221 ) and a coauthor of
another intermediate level C++ book I have collected the following
general rules of thumb:

Focus on your target audience. Don't be tempted to write advanced stuff
in a primer, for instance; Don't explain the obvious to advanced readers
either.

*Have several experts review your manuscripts; make sure they read every
line of text and code more than once.

*Compile every line of code -- you'd be surprised of the amount of typos
and errors a seemingly perfect program may contain.

*Be consistent. Use the same writing conventions, indentation, etc.
throughout the book.

*Talk to your editors frequently; hear their technical advice. They have
been involved in many similar projects before; their experience is
therefore invaluable. This however is not to say that you shouldn't
follow your intuition. Most importantly -- work with people you like and
trust.

*Post your manuscript on the Web as a beta version of your book to
elicit feedback from the public. Again, you'd be amazed of how many
errors and omissions your thoroughly proofread manuscript still
contains. My publisher does that with every new book; others do that
only occasionally.

*Meet your deadlines, always. Failing to meet one deadline causes a
landslide of delays.

>
>Assuming I can get a publisher interested in publishing my work, how much
am
>I likely to earn for my efforts, both up-front payment, and percentage
>royalty?

It's really hard to tell. Some big shots get around 20% of the net
income as royalties, others settle for 6 %. It's a matter of experience,
the risk level (is the book likely to be a bestseller?), the tentative
list price, the publisher's investment, the number of authors, the
book's topic, and -- your negotiation skills. As a beginner, consider
anything beyond 8% a reasonable bargain. As for the advance -- again
there are no rules here. You can get as little as $1000 or as much as
$75,000. Hear what your publisher offers and then try to negotiate.
However, focus your efforts on higher royalties rather than squeezing a
few more hundred bucks to the advance. Remember that an advance is
deducted from your royalties, after all.

>
>How much effort do you find that you as an author need to put in to getting
>a manuscript up to the standard required for publishing. i.e., around how
>many hours effort per page of printed material would you say is typical?

Much more than you ever thought it would take:-) Writing my book was a
non-stop 14X7 hours a week engagement for about 4 months + 2 more (a
little bit calmer) months for proofreading and editing. At the end, you
will recite whole chapters by heart:-) Forget about holidays, friends,
hobbies -- the book will take all your time. The schedules are always
tight because your publisher's competitors are probably producing their
own titles and you don't want your book to appear a month after one of
them came out with a book on the same subject.
Remember that you have a lot of things to take care of in addition to
the actual writing: answering reviewers' comments, checking your
editors' corrections (sometimes you have to rephrase complete pages),
making sure that the code appears in the correct font, compiling your
code again after you have fixed and changed it according to your
reviewers and editors' notes etc. etc.

This may sound a bit scary but I don't want you to get that impression.
The truth is that writing a book is one of the most rewarding
experiences in my life and if you don't have unrealistic dreams of
becoming rich, you'd gain an unparalleled feeling of accomplishment once
you get the chance to hold the first copy in your hand.

Danny Kalev

"persius" <mark.bower@persius.freeserve.co.uk> wrote:
>I'd like some advice on writing books for developers. Well, I can only pass
on my personal experience, which should be taken as such - a personal point
of view. Other authors might see it differently, although I know that most
authors I know in person would agree that they did not write their books
for making money. :-)

Regarding earnings from a publication: it's not worth it unless you happen
to write a best-seller. Practically any other occupation has a higher profit
rate in the IT business. Consulting is much more profitable. This is particularly
so in Germany, where I live, because of the taxation: you pay tax on the
earnings (royalties for instance), but can typically not deduct any expenses
that you have before the book is published.

Regarding the figures: Royalties, advances, and the like are subject to negotiations
with the publisher and vary accordingly.

Regarding effort: A coauthor and I spent approximately three years writing
a book of 600 pages, but we did it as a spare-time interest, definitely not
as a full-time occupation. It also drastically depends on how experienced
you are in writing. Today I am much faster than I used to be when I wrote
my first articles. It takes me a day (10 hours) to write a draft of a 4,000-word
article, which then goes through at least two review cycles with my co-author.
That's another 10 hours effort at least. So, it's 20-25 hours for 4,000
words. That's half a page per hour and is pretty fast.

And this is just writing. The initial research, putting together code samples,
figuring out the storyline and the didactic issues are not counted here and
typically take much longer than the actual typing. Take book reviews as an
example: first you read the entire book, more or less thoroughly, typically
look into other comparable books, discuss the book with a peer, figure out
the main points of the review, and eventually you write the actual 500-800
words that will be published, review and revise them, and often you also
have to do proofreading after editing and typesetting by the publisher.
Typing the review is a minor part of the effort. Plus you should not underestimate
the production process of a book publication: countless reviews, proofreading,
putting together the index, all of which is boring, tedious work that sums
up quite a bit.

Bottom line: I wouldn't write a book for making money. If that is the primary
motivation, I would strongly recommend against it.

Angelika Langer
Training & Consulting
Object-Oriented Software Development in C++ & Java