When the Web was created everyone trusted each other, mostly because everyone knew each other. The network was much smaller back then, and everyone used the network the same way. It was not the free-for-all it is today. That said, the underlying infrastructure of the network hasn’t changed all that much, but what is being exchanged over the network has changed. Today, people are managing their money, conducting business transactions, and hosting sensitive data over the Net.
The Internet still works fine as long as we trust each other. You know, that same kind of trust that lets us walk down the street, go to the store, or sing karaoke at the local bar without fear. In fact, without trust, you would never buy anything from Amazon or eBay again—let alone eat a hot dog.
Now, I don’t know about you, but I don’t trust everyone. I also want to keep my private data private and not let it leak out of my applications like motor oil from an old Buick. So, we must inspect the entire surface of the application and make sure the data stays in and the bad guys stay out. I start by asking myself how could data escape the system? Where can data be found or accessed? What security measures are currently in place to protect the data?
Some examples of where data leaks might occur are:
Runtime errors printed to the standard error or output stream. Depending on configuration, this information could be displayed to a system console or to an unprotected log file exposing details ...