Data Privacy and Protections

For the ongoing discussion of data theory and applications, Science News, an organization that reports research of various subjects, released an article called Theoretical computer science provides answers to data privacy problem. For the purpose of this article, concepts related to the handling of data are discussed, and their applications. Specifically, the purposes that relate data to people are recognizable. Science News (2015) said, “Some data may be trivial, but in many cases, data are deeply personal. They can even influence our insurance premiums or the price we pay for a product online.” An example of a working system model with privacy worked with was reported by Science News as researched by a professor of computer science in Harvard University and former director of the Center of Research on Computation and Society, Salil Vadhan (2015). Described as “differential privacy” (2015), this concept protects data with approximations. With this premise, a series of queries may provide an identifiable pattern; a summation may be reckoned. The calculus of this approximation would then be equivalent to an integral that will produce an exact result. Science News reported that this is defended against by good judgment that would be increasing randomization and taking full care in comparing characteristics across queries (2015). This assessment seems vague because what people view as good judgment depends on initial assumptions about reality.

Although rule utilitarianism, for example, is discussed in contemporary times as a serious ethical concept, there is debate about its efficacy. Bo Brinkman and Alton F. Sanders describe rule utilitarianism. Brinkman and Sanders (2013) said, “In rule utilitarianism, we select a set of rules, and each act is evaluated as to whether it conforms to them” (p. 16). In this case, a rule utilitarian approach might be that the supposedly good judgment of differential privacy is people who may have access will have the keys to accurate information retrieval, and people with incorrect keys will get incorrect values returned to them. In the late 20th century AD, Winslett, et al. (1994) supposed a similar proposal. Winslett et al. (1994) said, “We believe that many of the MLS problems can be resolved by directly addressing the question of what an MLS database means, rather than making syntactic adjustments to avoid semantic problems” (p. 627). This proposal essentially asserted that lying is an effective form of confidentiality for cyber security. Although the supposed truth of relative semantics makes sense for the access key holders for secret information in short term, this causes a failure of integrity. For a historical account, the Apostle Paul discussed the Hebrew Bible. Paul (KJV) said, “As it is written, There is none righteous, no, not one:” (Rom 3:10). This was the view before grace.

As Joachim Biskup showed, lying is not the correct approach. Biskup (2000) said, “The initial belief and the first k-1 answers (lies) would reveal the last secret psi_k.” This is a mathematical statement that reflects a Biblical prophecy. The Son of Man (KJV) said, “Therefore whatsoever ye have spoken in darkness shall be heard in the light; and that which ye have spoken in the ear in closets shall be proclaimed upon the housetops” (Lu 12:3). Therefore, when randomizing data for the purpose of producing lies or untrue information for those who seek unauthorized access to data, I think that more secure concepts require something other than this sort of concept because true information will eventually be revealed, though that may be understood as defense in depth with protections being more understood as unapproachable like the image that a security guard for a crucial represents to attackers, defenders, and other vulnerable groups. Similarly, an available mathematical problem that computers would have difficulty solving would serve as a known defense that would be well defensible in contemporary times.

References

Biskup, J. (2000). For unknown secrecies refusal is better than lying. Data & Knowledge Engineering, 33(1), 1-23. doi:10.1016/S0169-023X(99)00043-9

Brinkman, W. J., & Sanders, A. F. (2013). Ethics in a Computing Culture. Boston, MA: Cengage Learning.

Marianne Winslett, Kenneth Smith, and Xiaolei Qian. 1994. Formal query languages for secure relational databases. ACM Trans. Database Syst. 19, 4 (December 1994), 626-662. DOI=http://0-dx.doi.org.library.regent.edu/10.1145/195664.195675

National Science Foundation. (2015, October 7). Theoretical computer science provides answers to data privacy problem: New tools allow researchers to share and study sensitive data safely by applying ‘differential privacy’. ScienceDaily. Retrieved December 25, 2016 from http://www.sciencedaily.com/releases/2015/10/151007144933.htm

Advertisements

Computer Locks and Threads

For this post, an area of threads is discussed. On January the first of 2015, Davidlohr Bueso released a paper called Scalability Techniques for Practical Synchronization Primitives dealing with locks. According to Bueso, locks are a method of concurrent execution of more than one thread. The importance of these locks is that they provide security and accuracy for the context of execution through excluding mutually. Furthermore, the efficiency and accuracy of locks are dependent on hardware architecture. Bueso continued by explaining that the Linux kernel utilizes pointers with locks, an attribute of the C programming language that others such as Java and Visual Basic do not contain. The reasons for this are that the computer is aware of when to stop spinning, and as a debugging tool it allows for deadlock detection. As Bueso mentioned, previous strategies as old as 1975 conceptualized lock ownership within databases. This is useful for deciding for or against reentrant locks that usually utilize a counter field. There are multiple lock paths including Fastpath, Midpath, and Slowpath. All of these can block, the aforementioned primitives require a sleeping context for safety. These locks can be a determining factor in performance. Along with the previously mentioned information, there is a resource cost of utilizing a certain lock that is determined by the size and latency. Essentially, structure size determines CPU cache and memory resource allocation. This becomes increasingly relevant as a structure is utilized with increasing frequency. Fairness is determined by the strictness of semantics. Bueso concluded by explaining the cost of not optimizing with having initiative as greatly important.

While reading this, I thought of the Biblical Scriptures. The Bible has much to say about fairness and the cost of not caring for the resources that our blessed Father in heaven has given us. The Bible says, “The LORD God took the man and put him in the Garden of Eden to work it and take care of it” (NIV, Gen 2:15). The blessing of God is for man, male and female, to properly care for God’s creation. This can be interpreted as the bestowing of a responsibility and a privilege. This has helped me understand more about how computer hardware is directly affected by software. Particularly, strictness of semantics determines the prevention or allowing of resource starvation. In my personal experience, software applications starving other parts of a user experience from correct use can be burdensome, and efficiency of thread use may prevent that.

Bibliography

BUESO, D. (2015). Scalability Techniques for Practical Synchronization Primitives. Communications Of The ACM, 58(1), 66-74. doi:10.1145/2687882