#### Answer

The point here is to think about how software properties can be measured. One approach for estimating the number of errors in a piece of software is to intentionally place some errors in the software when it is designed. Then, after the software has supposedly been debugged, check to see how many of the original errors are still present. For example, if you intentionally place seven errors in the software and find that five have been removed after debugging, then you might conjecture that only 5⁄7 of the total errors in the software have been removed.