Distributed data usage control for web applications

Representation-Independent Data Usage Control

“…While we do not claim that our system cannot be circumvented, we have some confidence that a reasonable level of security can be attained [25,10]. Performance-wise, we currently are faced with an overhead of one to two orders of magnitude [3,4].…”

“…An example policy we can enforce is presented in Appendix B.2. Web Browser: Firefox A third instance of our model at the browser level extends an existing usage control extension for the Firefox web browser [25]. In this scenario, we want to protect sensitive web page content from malicious usage by the user of the browser.…”

“…A general protocol for such a communication among arbitrary parties is the subject of current work, so we hard-coded a communication solution tailored for this specific scenario: we made the Firefox monitor able to instruct the OS and X11 monitors about new policies and data flows from the browser level to the operating system and the windowing system, respectively. For our example, we adopted an existing solution [25] for policy retrieval and hard-coded inter-layer data flow observations.…”

“…We consider a social network use case [25] where a user watches a picture on someone else's profile page. Since the picture is considered sensitive, its usage is restricted.…”

“…opened, (lines 12-15) at most 4 further times and within 30 seconds (1 timestep = 1 second) after the first use (lines [8][9][10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27]; further attempts of opening the file will result in opening a predefined error message (lines 28-34).…”