Software protection

A computer system's security can be compromised in many ways—a denial-of-service attack can make a server inoperable, a worm can destroy a user's private data, or an eavesdropper can reap financial rewards by inserting himself in the communication link between a customer and her bank through a man-in-the-middle (MITM) attack. What all these scenarios have in common is that the adversary is an untrusted entity that attacks a system from the outside—we assume that the computers under attack are operated by benign and trusted users. But if we remove this assumption, if we allow anyone operating a computer system—from system administrators down to ordinary users—to compromise that system's security, we find ourselves in a scenario that has received comparatively little attention.

Enhancing Security Requirements Engineering by Organisational Learning

More and more software projects today are security-related in one way or the other. Requirements engineers often fail to recognise indicators for security problems which is a major source of security problems in practice. Identifying security-relevant requirements is labour-intensive and errorprone. In order to facilitate the security requirements elicitation process, we present an approach supporting organisational learning on security requirements by establishing company-wide experience resources, and a socio-technical network to benefit from them. The approach is based on modelling the flow of requirements and related experiences. Based on those models, we enable people to exchange experiences about security-requirements while they write and discuss project requirements. At the same time, the approach enables participating stakeholders to learn while they write requirements. This can increase security awareness and facilitate learning on both individual and organisational levels. As a basis for our approach, we introduce heuristic assistant tools which support reuse of existing security-related experiences. In particular, they include Bayesian classifiers which issue a warning automatically when new requirements seem to be security-relevant. Our results indicate that this is feasible, in particular if the classifier is trained with domain specific data and documents from previous projects. We show how the ability to identify security-relevant requirements can be improved using this approach. We illustrate our approach by providing a step-by-step example of how we improved the security requirements engineering process at the European Telecommunications Standards Institute (ETSI) and report on experiences made in this application.