Is software developed and tested different in military and government than in the commercial sector? Certainly. Is the outcome different? Not substantially. The cases studies we have examined for both circumstances show that successes and failures exist in every arena. Everyone recognizes that safety-critical systems need to be tested and examined whenever they are deployed, the only question is how much weight is given to this consideration, and government projects seem just as likely to weigh it too little as those in industry. The methodologies used by the government are different from those used by the commercial sector; but these differences far exceed the software arena, and do not seem to affect it excessively.

Without doubt, writing reliable software is one of the hardest tasks a programmer faces. Neither the commercial nor government sector routinely produces more reliable systems than the other, but when reliable systems do emerge, they are those that were well designed, well implemented and well tested. Software is not an area that can be short-changed for a computer system to be reliable. But even the most meticulous care cannot eliminate all bugs. There is no known no way to produce 100% bug-free code, and anyone implementing a critical system must be aware of this, no matter what the application and no matter how many precautions they take. In the end, any software system can fail

.