The Software Conspiracy: Why Software Companies Put Out Faulty Products, How They Can Hurt You, and What You Can Do About It: What You Don't Know ... and How It's Taking Control of Your Life [Englisch] [Gebundene Ausgabe]

Anyway, on to the content: I found the author made very interesting points so far, but few people buy the NASA reliability argument any more. I don't buy the Motorola argument either. Sure, reliable cell phones with poor usability and poor user interface, there obviously was a tradeoff to get that "quality". The writing style itself is good. There are a few missteps so far - like regression testing is not about improving forecasting through feedback, but simply testing changes against all past tests to ensure no new bugs have been introduced.

I think the fatal flaw in the book is that all software isn't the same. Software most consumers use is different from software that web sites use, is different from NASA, is different from medical equipment. They all vary in spec and in cost, deadline, features, and desirability.

Especially in business, if you wanted a spec, you and your offspring and probably even the company would be dead before the spec was written. In a changing environment, a better strategy is to develop in an evolving style, trying out things and fixing bugs as you go along. The users understand that often they don't know how to improve their jobs, they just want it easier and more reliable. You can only learn how to build what users need (not what they want - another flaw in the book) by building the software in iterations. Sure, this isn't shrink-wrap software, but then again, most programmers don't actually write shrink-wrap software, so he's already complaining about a minority of programmers. Programmers today are just as likely to be designing business processes, workflows and management consulting as they are simply coding.

A worthwhile read, but the book is just like the software it complains about, "good enough." If that didn't actually contradict the goals of the book, I'd give it five stars.

The central message of the book is that consumer software stinks because you, the consumer, buy it anyway and don't hold vendors responsible. Software intended for the general public has been so lousy for so long that people have become accustomed to it, even to the point of making apologies for the vendors. The author gives a reasonable amount of background on several aspects of software quality (technical, legal, business), some steps to take, and mentions a few resources.

The book is more about raising the level of awareness of the issue than about helping you deal with any particular bug though. Nothing will happen until there is a major change in public opinion.

Kudos for illustrating how the semi-technical press (PC Magazine and the like) contribute to the problem.

Although the author has the big picture right there are too many details wrong in his superficial presentation of the software development process for me to recommend the book enthusiastically. Nonetheless much that is in this book will probably be a revelation to the general public and so what really matters is that the big picture is right.

The resource section could have been a lot more directly useful, and should have a web site backing it. But there are some useful pointers.

So why do we allow the software companies to perpetuate this?

I for one am sick of it. The EUA should be abolished. We should not have to waive our rights to use a software. Especially after we purchased it, and we cannot return it once opened. If we were allowed to return a software after we used it and found it to be buggy, then this might be somewhat more acceptable.

There has to be some serious changes in this industry. As harddrive and memory increases in size, coders are lead to believed they can bloat the software as much as they want because users wouldn't be worried about space. But the more bloated software are, the more chances there are for conflicts and bugs.