Search for: All records

Many systems are built around the assumption that one ac- count corresponds to one user. Likewise, password creation and management is often studied in the context of single-user accounts. However, account and credential sharing is com- monplace, and password generation has not been thoroughly investigated in accounts shared among multiple users. We examine account sharing behaviors, as well as strategies and motivations for creating shared passwords, through a census- representative survey of U.S. users (n = 300). We found that password creation for shared accounts tends to be an individ- ual, rather than collaborative, process. While users tend to have broadly similar password creation strategies and goals for both their personal and shared accounts, they sometimes make security concessions in order to improve password us- ability and account accessibility in shared accounts. Password reuse is common among accounts collectively shared within a group, and almost a third of our participants either directly reuse or reuse a variant of a personal account password on a shared account. Based on our findings, we make recommen- dations for developers to facilitate safe sharing practices. 

During the Second World War, estimates of the number of tanks deployed by Germany were critically needed. The Allies adopted a successful statistical approach to estimate this information: assuming that the tanks are sequentially numbered starting from 1, if we observe k tanks from an unknown total of N, then the best linear unbiased estimator for N is M(1+1/k)-1 where M is the maximum observed serial number. However, in many situations, the original German Tank Problem is insufficient, since typically there are l > 1 factories, and tanks produced by different factories may have serial numbers in disjoint ranges that are often far separated.Clark, Gonye and Miller presented an unbiased estimator for N when the minimum serial number is unknown. Provided one identifies which samples correspond to which factory, one can then estimate each factory's range and summing the sizes of these ranges yields an estimate for the rival's total productivity. We construct an efficient procedure to estimate the total productivity and prove that it is effective when log l/log k is sufficiently small. In the final section, we show that given information about the gaps, we can make an estimator that performs orders of magnitude better when we have a small number of samples.