Privacy-protected microdata are often the desired output of a differentially private algorithm since microdata is familiar and convenient for downstream users. However, there is a statistical price for this kind of convenience. We show that an uncertainty principle governs the trade-off between accuracy for a population of interest (``sum query'') vs. accuracy for its component sub-populations (``point queries''). Compared to differentially private query answering systems that are not required to produce microdata, accuracy can degrade by a logarithmic factor. For example, in the case of pure differential privacy, without the microdata requirement, one can provide noisy answers to the sum query and all point queries while guaranteeing that each answer has squared error $O(1/\epsilon^2)$. With the microdata requirement, one must choose between allowing an additional $\log^2(d)$ factor ($d$ is the number of point queries) for some point queries or allowing an extra $O(d^2)$ factor for the sum query. We present lower bounds for pure, approximate, and concentrated differential privacy. We propose mitigation strategies and create a collection of benchmark datasets that can be used for public study of this problem.

Potentially dangerous cryptography errors are well-documented in many applications. Conventional wisdom suggests that many of these errors are caused by cryptographic Application Programming Interfaces (APIs) that are too complicated, have insecure defaults, or are poorly documented. To address this problem, researchers have created several cryptographic libraries that they claim are more usable, however, none of these libraries have been empirically evaluated for their ability to promote more secure development. This paper is the first to examine both how and why the design and resulting usability of different cryptographic libraries affects the security of code written with them, with the goal of understanding how to build effective future libraries. We conducted a controlled experiment in which 256 Python developers recruited from GitHub attempt common tasks involving symmetric and asymmetric cryptography using one of five different APIs. We examine their resulting code for functional correctness and security, and compare their results to their self-reported sentiment about their assigned library. Our results suggest that while APIs designed for simplicity can provide security benefits - reducing the decision space, as expected, prevents choice of insecure parameters - simplicity is not enough. Poor documentation, missing code examples, and a lack of auxiliary features such as secure key storage, caused even participants assigned to simplified libraries to struggle with both basic functional correctness and security. Surprisingly, the availability of comprehensive documentation and easy-to-use code examples seems to compensate for more complicated APIs in terms of functionally correct results and participant reactions, however, this did not extend to security results. We find it particularly concerning that for about 20% of functionally correct tasks, across libraries, participants believed their code was secure when it was not. Our results suggest that while new cryptographic libraries that want to promote effective security should offer a simple, convenient interface, this is not enough: they should also, and perhaps more importantly, ensure support for a broad range of common tasks and provide accessible documentation with secure, easy-to-use code examples.

Due to lack of mature techniques in privacy-preserving information retrieval (IR), concerns about information privacy and security have become serious obstacles that prevent valuable user data to be used in IR research such as studies on query logs, social media, and medical record retrieval. In SIGIR 2014 and SIGIR 2015, we have run the privacy-preserving IR workshops exploring and understanding the privacy and security risks in information retrieval. This year, we continue the efforts of connecting the two disciplines of IR and privacy/security by organizing this workshop. We target on three themes, differential privacy and IR dataset release, privacy in search and browsing, and privacy in social media. The workshop includes panels with researchers from both fields on these three themes, as well as invite industry speakers for real-world challenges. The goals of this workshop include (1) bringing together the two research fields, and (2) yielding fruitful collaborations.

Bulk data analysis eschews file extraction and analysis, common in forensic practice today, and instead processes data in “bulk,” recognizing and extracting salient details (“features”) of use in the typical digital forensics investigation. This article presents the requirements, design and implementation of the bulk_extractor, a high-performance carving and feature extraction tool that uses bulk data analysis to allow the triage and rapid exploitation of digital media. Bulk data analysis and the bulk_extractor are designed to complement traditional forensic approaches, not replace them. The approach and implementation offer several important advances over today's forensic tools, including optimistic decompression of compressed data, context-based stop-lists, and the use of a “forensic path” to document both the physical location and forensic transformations necessary to reconstruct extracted evidence. The bulk_extractor is a stream-based forensic tool, meaning that it scans the entire media from beginning to end without seeking the disk head, and is fully parallelized, allowing it to work at the maximum I/O capabilities of the underlying hardware (provided that the system has sufficient CPU resources). Although bulk_extractor was developed as a research prototype, it has proved useful in actual police investigations, two of which this article recounts.

This paper presents a novel solution to the problem of determining the ownership of carved information found on disk drives and other storage media that have been used by more than one person. When a computer is subject to forensic examination, information may be found that cannot be readily ascribed to a specific user. Such information is typically not located in a specific file or directory, but is found through file carving, which recovers data from unallocated disk sectors. Because the data is carved, it does not have associated file system metadata, and its owner cannot be readily ascertained. The technique presented in this paper starts by automatically recovering both file system metadata as well as extended metadata embedded in files (for instance, embedded timestamps) directly from a disk image. This metadata is then used to find exemplars and to create a machine learning classifier that can be used to ascertain the likely owner of the carved data. The resulting classifier is well suited for use in a legal setting since the accuracy can be easily verified using cross-validation. Our technique also results in a classifier that is easily validated by manual inspection. We report results of the technique applied to both specific hard drive data created in our laboratory and multiuser drives that we acquired on the secondary market. We also present a tool set that automatically creates the classifier and performs validation.

No abstract available.

Much effort has been expended in recent years to create large sets of hash codes from known files. Distributing these sets has become more difficult as these sets grow larger. Meanwhile the value of these sets for eliminating the need to analyze "known goods'' has decreased as hard drives have dramatically increased in storage capacity. This paper evaluates the use of bloom filters (BFs) to distribute the National Software Reference Library's (NSRL) Reference Data Set (RDS)version 2.19, with 13 million SHA-1 hashes. We present an open source reference BF implementation and validate it against a large collection of disk images. We discuss the tuning of the filters, evaluate discuss how they can be used to enable new forensic functionality, and present a novel attack against bloom filters.

We present an integrated security model for a low-cost laptop that will be widely deployed throughout the developing world. Implemented on top of Linux operating system, the model is designed to restrict the laptop's software without restricting the laptop's user.

Security toolbars in a web browser show security-related information about a website to help users detect phishing attacks. Because the toolbars are designed for humans to use, they should be evaluated for usability -- that is, whether these toolbars really prevent users from being tricked into providing personal information. We conducted two user studies of three security toolbars and other browser security indicators and found them all ineffective at preventing phishing attacks. Even though subjects were asked to pay attention to the toolbar, many failed to look at it; others disregarded or explained away the toolbars' warnings if the content of web pages looked legitimate. We found that many subjects do not understand phishing attacks or realize how sophisticated such attacks can be.

Cryptographically protected email has a justly deserved reputation of being difficult to use. Based on an analysis of the PEM, PGP and S/MIME standards and a survey of 470 merchants who sell products on Amazon.com, we argue that the vast majority of Internet users can start enjoying digitally signed email today. We present suggestions for the use of digitally signed mail in e-commerce and simple modifications to webmail systems that would significantly increase integrity, privacy and authorship guarantees that those systems make. We then show how to use the S/MIME standard to extend such protections Internet-wide. Finally, we argue that software vendors must make minor changes to the way that mail clients store email before unsophisticated users can safely handle mail that is sealed with encryption.

Secure email has struggled with signifcant obstacles to adoption, among them the low usability of encryption software and the cost and overhead of obtaining public key certificates. Key continuity management (KCM) has been proposed as a way to lower these barriers to adoption, by making key generation, key management, and message signing essentially automatic. We present the first user study of KCM-secured email, conducted on naïve users who had no previous experience with secure email. Our secure email prototype, CoPilot, color-codes messages depending on whether they were signed and whether the signer was previously known or unknown. This interface makes users signicantly less susceptible to social engineering attacks overall, but new-identity attacks (from email addresses never seen before) are still effective. Also, naïve users do use the Sign and Encrypt button on the Outlook Express toolbar when the situation seems to warrant it, even without explicit instruction, although some falsely hoped that Encrypt would protect a secret message even when sent directly to an attacker. We conclude that KCM is a workable model for improving email security today, but work is needed to alert users to "phishing" attacks.