Now I've panned an Elfriede Dustin book before and then enjoyed another one. So thus far I have a 50/50 success rate so I pulled Effective Software Testing from the shelf, curious as to how I would react.

[amazon.com][amazon.co.uk]

You can see the list of topics covered on the PersonEd site.

I made a whole bunch of nitpicky wee notes when I went through the book which I don't really want to write down here as it would ultimately serve no purpose since I can not recommend this book. But I do recommend that you review the preview available from Google Books and make up your own mind.

Most of my notes relate to the content not having 'specific' enough information. Mildly ironic given the book's subtitle "50 specific ways to improve  your testing".

You can see from the topics that Elfriede Dustin wants to present the topics in the order of a 'typical' life cycle.

I found that I agreed with the general intent behind statements made in the book, but not the whole statement e.g. I can agree with "testers need deeper knowledge" (of requirements) and that "understanding the thought process used during the specification of product functionality" can help testers. But I didn't agree with "tester need deeper knowledge that can only come from understanding the though process used during the specification of product functionality."  Admittedly that seems like a pretty specific statement but not one that helps me much if, as a tester, I don't get involved early - what else could I do? The book doesn't tell me.

Too often I found the book telling me that 'some action' will help 'with test planning' - but not how. So if a junior tester reads this then they will learn 'I need to have X' but not necessarily how to use it when they get it.

I liked the requirements section starting with a quote from Christopher Alexander. But the requirements section itself could have made its way into any general software development book or business analysis book. Not enough specific testing information.

So I hoped for more from this book. And the title didn't deliver on its promise to me.

The book ended up reading like another introductory book on Software Testing. But a fairly general one at that.

As an example of where it let me down. A useful section on "Use Logging to Increase System Testability" but again overly general and didn't go far enough. A more specific book, would not just have told me to review 'items written to the log file' it would have told me about GREP, or TAIL, or REGEX for pulling out errors, or a quick Perl script to filter the results, or... something specific I could do with log files to make an immediate improvement in my test process. In its current form, the chapter will help junior testers who had never considered that they could ask for a log file to get written. But not much more.

Sadly then, having ploughed through it, I can't recommend this book but I don't think I represent the target audience so if you remain curious about the book then fortunately you can check the preview available from Google Books and make up your own mind. Which I recommend you do as other people have obviously received the book well and you may find value in it.

Related Links:

• amazon.com
• amazon.co.uk
• PersonEd contents list
• preview available from Google Books
• Elfriede Dustin InformIT page
• Interview with Elfriede Dustin
• a review by Gary Pollice on IBM developer works