Mike Bursell

Trust in Computer Systems and the Cloud


Скачать книгу

the likelihood of the keyboard failing is 4%, or 0.04, whereas the likelihood of the monitor failing is only 1%, or 0.01. If I were to consider this information on its own, it would seem that I should worry more about the keyboard than the monitor, until I take into account the cost of replacement: the keyboard would cost me $15 to replace, whereas the monitor would cost me $400 to replace. We have the following risk calculations then:

      It turns out that if I care about risk, I should be more concerned about the monitor than the keyboard. Once we have calculated the risk, we can then consider mitigations: what to do to manage the risk. In the case of my desktop computer, I might decide to take out an extended manufacturer's warranty to cover the monitor but just choose to buy a new keyboard if that breaks.

      Risk is all around us and has been since before humans became truly human, living in groups and inhabiting a social structure. We can think of risk as arising in four categories:

ASSESSMENT MITIGATION
Easy If there are predators nearby, they might kill us … Easy … so we should run away or hide.
Easy If our leader gets an infection, she may die … Difficult … but we don't know how to avoid or effectively treat infection.
Difficult If the river floods, our possessions may be washed away … Easy … but if we camp farther away from the river, we are safer.
Difficult If I eat this fruit, it may poison me … Difficult … but I have no other foodstuffs nearby and may go hungry or even starve if I do not eat it.

      For the easy-to-assess categories, both the probability and the loss are simple to calculate. For the difficult-to-assess categories, either the probability or the loss is hard to calculate. What is not clear from the simple formula we used earlier to calculate risk is that you are usually calculating a risk against something that is generally a benefit. In the case of the risk associated with the river, there are advantages to camping close to it—easy access to water and ability to fish, for example—and in the case of the fruit, the benefit of eating it will be that it may nourish me, and I do not need to trek further afield to find something else to eat, thereby using up valuable energy.

      Many of the risks associated with interacting with other humans fit within the last category: difficult to assess and difficult to mitigate. In terms of assessment, humans often act in their own interests rather than those of others, or even of a larger group; and the impact of an individual not cooperating may be small—hurt feelings, for example—or large—inability to catch game—or even retribution towards a member of the group. In terms of mitigation, it is often very difficult to guess what actions to take to encourage an individual, particularly one you do not already know, to ensure that they interact with you in a positive manner. You can, of course, avoid any interactions at all, but that means you lose access to any benefits from such interactions, and those benefits can be very significant: new knowledge, teamwork for hunting, more strength to move objects, safety in numbers, even having access to a larger gene pool, to name just a few.

      Risk is important in the world of IT and computing. Organisations need to know whether their systems will work as expected or if they will fail for any one of many reasons: for example, hardware failure, loss of power, malicious compromise, poor software. Given that trust is a way of mitigating risk, are there opportunities to use trust—to transfer what humans have learned from creating and maintaining trust relationships—and transfer it to this world? We could say that humans need to “trust” their systems. If we think back to the cases presented earlier in the chapter, this fits our third example, where we discussed the bank trusting its IT systems.

      Defining Trust in Systems

      There is a growing corpus of research and writing around how humans build trust relationships to each other and to organisations, and this is beginning to be applied to how humans and organisations trust computer systems. What is missing is often a realisation that interactions between computer systems themselves—case four in our earlier examples—are frequently modelled in terms of trust relationships. But as these models lack the rigour and theoretical underpinnings to allow strong statements to be made about what is really going on, we are left without the ability to allow detailed discussion of risk and risk mitigation.

      Why does this matter, though? The first answer is that when you are running a business, you need to know that all the pieces are correct and doing the correct thing in relationship to each other. This set of behaviours and relationships makes up a system, and the pieces its components, a subject to which we will return in Chapter 5: The Importance of Systems. We can think of this as similar to ensuring that your car is made up of the correct parts, placed in the correct locations. If you have the wrong brake cable, then