Is any one here familiar with Kevin Leininger's book "Solaris Developer's
Tool Kit"? If so, what are your thoughts about it? (I know I could just
order it online, but I'm trying to save my cash! If need be, I will do
Also, what material would you like to see in a programming book that
has a Solaris bias?
Why am I asking these questions? Some publishers have approached me with
the idea of writing a C programming book with a Solaris bias, so I'm checking
out the competition, and soliciting ideas for material to include. I'm
envisioning a reference level type work, in much the same way that Rich
Stevens' UNIX Network Programming is THE network programming reference.
NT tries to do almost everything UNIX does, but fails - miserably.
The use of Windoze cripples the mind; its use should, therefore, be
regarded as a criminal offence. (With apologies to Edsger W. Dijkstra)
I guess you will be competing with Stevens' Advanced Programming in the
Unix Environment, so you need to think how you can improve on that.
If you are looking for a Solaris bias, you could think about issues
which are more Solaris-specific and not covered by Stevens' book (i.e.
complimentary rather than competing).
An area which isn't so well covered in books I've scanned through is how
to design software which scales well on Solaris multi-processor systems.
I've seen some projects which made pretty fatal mistakes at the beginning
such that later redesign was necessary. With appropriate initial guidance,
that could easily have been avoided.
Consultant Software Engineer
in this age of Java programming, many people are using Java
threads and not any other platform specific threads API (such
as solaris threads) or even a C based standard API such as pthreads.
Java, in a way, makes platform specific programming knowledge
not as important as before (unless one is doing system level
type development of course).
> I hadn't heard of it before, but I just looked it up on amazon.com. It
> seems that this book was published in 1995! Pretty old in this field,
> wouldn't seem like much of competitor. Your book is going to offer
> more up-to-date information. Go for it!
Thanks for the encouragement! The proposal is nearly ready to send off
to the publishers - then it's outta my hands!
> I guess you will be competing with Stevens' Advanced Programming in the
> Unix Environment, so you need to think how you can improve on that.
Yes, that excellent book will be my main competition. Good as it is,
some of the material is a bit dated now.
> If you are looking for a Solaris bias, you could think about issues
> which are more Solaris-specific and not covered by Stevens' book (i.e.
> complimentary rather than competing).
The book I'm planning will almost fit both categories. Hopefully, being
more up to date, and a Solaris bias will be two good selling points from
both points of view. If you already have APUE, a lot of my material will
be less important (to be the reference I envisage, some overlap is
inevitable), but it will make an ideal complement to APUE. And if you
don't already have Rich's book, then maybe my book would be a better choice.
> An area which isn't so well covered in books I've scanned through is how
> to design software which scales well on Solaris multi-processor systems.
> I've seen some projects which made pretty fatal mistakes at the beginning
> such that later redesign was necessary. With appropriate initial guidance,
> that could easily have been avoided.
I will be covering multi-threaded programming. I don't intend my book to
be a "how to program" type book, but I think that a section or two about
scalable design would be a good thing to include.
Thanks for your thoughts,