Title: RESULTS

Results

Trusted computing can also be used for a reverse digital rights management. Instead of protecting digital contents of service providers, the reverse deployment of Trusted Computing in combination with a monitor protects personal data according to the agreed processing rules between a user and a service provider. Service providers show their trustworthiness by using a certain monitor which observes the activities of storing and delegating personal data. Trusted computing is thereby used for the attestation of using such a monitor. The deployment of Trusted Computing as specified by the Trusted Computing Group may therefore be used with minor modifications. One necessary modification of the TCG specification is to close the time gap between the attestation of a service application (monitor) and the collection of personal data, since an information system can be modified between these two activities. 

This deliverable has shown that the protection of personal data by enforcing agreed privacy policies is done by using TC-attestated service access points and a monitor in combination with mechanisms of information flow analysis. By using TC-attestated service access points, a user can be sure that his personal data will indeed be sent to the attestated service application. The combination with a monitor detects an undesired storing and delegation of personal data. 

There are more straightforward ways to implement privacy-preserving information architectures, e.g. by utilizing a centralized architecture in which the privacy-preserving provider-side functionality is realized as trusted software based on Trusted Computing. However, these approaches seem to be unsuitable because they are far less generic: Whenever some part of the respective software is patched, upgraded or replaced, the entire system has to be analyzed again in order to determine its trustworthiness, a process that is problematic in itself due to its complexity. In the described solution of section only a comparatively small part of the overall system is based on Trusted Computing. Because agent platforms can be utilized for a large variety of tasks, and because Trusted Computing seems to be the most promising approach to realize secure and trusted agent environments, it seems reasonable to assume that the respective mechanisms will be generally available in the future, independent of specific solutions such as the one described here.