Tonight while I was handing out candy on Halloween as the children came to the door trick-or-treating getting whatever candy I’ve not yet eaten. I started thinking about scary stories the security universe. Some of the things we do in Security could be compared to the old fable of the cursed monkey’s paw, which is one of my favorite stories.
For those who don’t know what this story is, the quick version of the story is essentially there is a monkey’s paw, an actual severed appendage of a monkey (it’s not some sort of figurative item). It has some fingers on it that may or may not signify the number of wishes used. The paw is indestructible, the previous owner doesn’t want it, but can’t get rid of it until some unsuspecting suckers shows up. The idea is you make a wish you get three wishes or five or whatever depending upon the version of the story that’s told (these old folk tales can differ greatly depending on what part of the world is telling them) and then the monkey paw gives you exactly what you asked for. The problem is what you asked for comes with horrifying consequences. For example there was an old man who had the paw and he asked for $200, the next day he got his $200 because his son was killed at work and they brought him $200 of his last paycheck. Of course there’s different variants of this but the basic idea is the paw seems clever, it grants wishes, but every wish comes with terrible consequences.
This story got me thinking about security, how we ask questions and how we answer questions. What if we think about this in the context of application security specifically for this example. If someone was to ask the security the question “does this code have a buffer overflow in it?” The person I asked for help is going to look for buffer overflows and they may or may not notice that it has a SQL injection problem. Or maybe it has an integer overflow or some other problem. The point is that’s not what they were looking for so we didn’t ask the right question. You can even bring this little farther and occasionally someone might ask the question “is my system secure” the answer is definitively no. You don’t even have to look at it to answer that question and so they don’t even know what to ask in reality. They are asking the monkey paw to bring them their money, it’s going to do it, but they’re not going to like the consequences.
So this really makes me think about how we frame the question since the questions we ask are super important, getting the details right is a big deal. However there’s also another side to asking questions and that’s being the human receiving the question. You have to be rational and sane in the way you deal with the person asking those questions. If we are the monkey’s paw; only giving people the technical answer to the technical question, odds are good we aren’t actually helping them.
As I sit here on this cold windy Halloween waiting for the kids to come and take all the candy that I keep eating, it really makes me think: as security practitioners we need to be very mindful of the fact that the questions people are asking us might not really be the answers they want. It’s up to us as humans, rather than monkey paws, to interpret the intent behind the person, what is the question they really want to ask, then give them answers they can use, answers they need, and answers that are actually helpful.