Now, some environmental factors that we need to consider. One is the attack surface evaluation. Now here, we need to validate that what we imagine or suspect might be our attack surface is in fact that or make the determination of what it actually is. Now, the surface of the area that we're examining typically represents a spot or multiple places where authenticated users can't run or input code into the system that we're testing, and thus extract data from the system.

Now, the security testing that we've been doing is about confirming resiliency with the intention to discover if the software can be made to fail or permit or even enable exploitation. Now, the security testing validates the presence and effectiveness of the security controls in the test artifact or in its environment. And through this, we are able to discern what the attack surface is, validate what is genuinely a problem in the attack surface. And that should lead to a program of shrinking the attack surface to the smallest feasible attack surface that we can.

Now, in the test artifacts, our environmental factors to be considered are, never to use actual live data or live modules. Now, of course, this can pose a bit of a problem, or at least a puzzle, because what we're trying to do is work with something that simulates real conditions as nearly as possible. So we need to use something that authentically represents real world or live operations, but we can never use the actual live data itself, because anything less, we would have to question, whether or not we were given true representations of cause, effect and behaviors that we would encounter in the real environment?

To do this, we have to create proper test data and there is a life cycle that is involved in this. The artifacts that we're using as the test objects must be protected like everything else in our project to make sure that they stay authentic, and that they will produce, every time that they're used, reliable results so that we can be sure that what we're working with is a true and fair representation of the conditions.

Now, it has to be looked at in the same way that any other craftsman, for the people doing this such as yourself are indeed craftsman. We need to be sure that we calibrate or sharpen our tools and test data is certainly one of the most important aspects of this tool, sharpening. We have to protect it from any modifications, not put under proper configuration management. If we test with inaccurately or poorly built data, our results are going to be equally poor. So it needs to be properly constructed and then protected very much like the actual operational system in its data would be.