Comments on: Tracking Website Data-Collection and Privacy Practices with the iWatch Web Crawler http://usablesecurity.com/2007/07/19/tracking-website-data-collection-and-privacy-practices-with-the-iwatch-web-crawler/ Every system has a user. Thu, 20 Nov 2008 22:46:32 +0000 http://wordpress.org/?v=abc By: Chandan Sarkar http://usablesecurity.com/2007/07/19/tracking-website-data-collection-and-privacy-practices-with-the-iwatch-web-crawler/#comment-68386 Chandan Sarkar Sun, 29 Jul 2007 17:45:08 +0000 http://usablesecurity.com/2007/07/19/tracking-website-data-collection-and-privacy-practices-with-the-iwatch-web-crawler/#comment-68386 The interesting part of the graph is that it is a directed graph. An Edge from vertex A to B means that A writes a cookie that only B can read more specifically A is the cookie writer and B the cookie reader. So what is more relevant for the importance of a node: Its indegree (how many cookies are written in other sites that can be read by this site.) and Its outdegree (How many cookies does this site write in the name of third party). Anotther two important feature of the graph are 1) RandomWalkBetweenness which specifies based on random walks, measuring the expected number of times a node is traversed by a random walk averaged over all pairs of nodes, and number of HITS which calculates the "hubs-and-authorities" importance measures for each node in a graph. The interesting part of the graph is that it is a directed graph. An Edge from vertex A to B means that A writes a cookie that only B can read more specifically A is the cookie writer and B the cookie reader. So what is more relevant for the importance of a node: Its indegree (how many cookies are written in other sites that can be read by this site.) and Its outdegree (How many cookies does this site write in the name of third party).

Anotther two important feature of the graph are
1) RandomWalkBetweenness which specifies based on random walks, measuring the expected number of times a node is traversed by a random walk averaged over all pairs of nodes, and number of
HITS which calculates the “hubs-and-authorities” importance measures for each node in a graph.

]]> By: Andrew http://usablesecurity.com/2007/07/19/tracking-website-data-collection-and-privacy-practices-with-the-iwatch-web-crawler/#comment-66707 Andrew Thu, 19 Jul 2007 17:49:22 +0000 http://usablesecurity.com/2007/07/19/tracking-website-data-collection-and-privacy-practices-with-the-iwatch-web-crawler/#comment-66707 Figure 9 in the paper, which was described as a teaser for future work, is interesting. Apparently it shows the connections between web sites based on third party cookies. It provokes interesting thoughts about privacy because it could be an illustration of the potential data aggregation between different domains. It should be of interest to people studying the issues/opportunities of <a href="http://searchenginewatch.com/showPage.html?page=3605361" rel="nofollow">behavioral advertising</a>. Figure 9 in the paper, which was described as a teaser for future work, is interesting. Apparently it shows the connections between web sites based on third party cookies.

It provokes interesting thoughts about privacy because it could be an illustration of the potential data aggregation between different domains. It should be of interest to people studying the issues/opportunities of behavioral advertising.

]]>