Most Popular Books by Allison Sullivan

The lives and deaths of the famous hold a special intrigue for people worldwide. In the Funerals of the Famous series, you''ll find stories ofsome of the most memorable personalities in history.

Abstract: One basic goal for any pediatric occupational therapist is to provide each child who receives services with the most effective treatment program possible. The treatment process should begin with the therapist performing an evaluation of the client in which two fundamental aspects of the child''s life are examined: the functional skills that the child is able to utilize in performing activities of daily living, and the child''s level of independence, or the amount of assistance required by the child in order to perform these tasks. Unfortunately there are few standardized instruments which can be used by pediatric occupational therapists for evaluation purposes (Campbell, 1989). Additionally, there is little clinical data that can provide therapists with a consistent definition of those components which truly constitute "independence", and it is not known whether the clinical definition of this concept should vary from population to population. In other words, the question should be raised as to whether or not those components which describe "independence" for disabled and non-disabled children include the same, similar, or different concerns. The Pediatric Evaluation of Disability Inventory (PEDI) is the first functional assessment designed specifically for the purpose of measuring a child''s functional skill level combined with an assessment of that child''s level of independence. This test records children''s functional performance in a variety of daily living skills and social behaviors. It is the opinion of the investigators that a relationship exists between functional ability, as measured by the Functional Skills/Behaviors section of the PEDI (Part One), and level of independence, as measured by the Caregiver Assistance/Modifications section of the PEDI (Part Two). This investigation will analyze the nature of this relationship. It is our belief that if the PEDI can describe the relationship between functional skill and independence accurately, this concept can be used to better evaluate and assess the needs of the clients who receive pediatric occupational therapy services.

Writing declarative models of software designs and analyzing them to detect defects is an effective methodology for developing more dependable software systems. However, writing such models correctly can be challenging for practitioners who may not be proficient in declarative programming, and their models themselves may be buggy. We introduce the foundations of a novel test automation framework, AUnit, which we envision for testing declarative models written in Alloy -- a first-order, relational language that is supported by its SAT-based analyzer. We take inspiration from the success of the family of xUnit frameworks that are used widely in practice for test automation, albeit for imperative or object-oriented programs. The key novelty of our work is to define a basis for unit testing for Alloy, specifically, to define the concepts of test case and test coverage as well as coverage criteria for declarative models. We reduce the problems of declarative test execution and coverage computation to partial evaluation without requiring SAT solving. Our vision is to blend how developers write unit tests in commonly used programming languages with how Alloy users formulate their models in Alloy, thereby facilitating the development and testing of Alloy models for both new Alloy users as well as experts. We illustrate our ideas using a small but complex Alloy model. While we focus on Alloy, our ideas generalize to other declarative languages (such as Z, B, ASM).

Models of software systems, e.g., designs, play an important role in the development of reliable and dependable systems. However, writing correct designs is hard. What makes formulating desired design properties particularly hard is the common lack of intuitive and effective techniques for validating their correctness. Despite significant advances in developing notations and languages for writing designs, techniques for validating them are often not as advanced and pose an undue burden on the users. Our thesis is that some foundational and widely used techniques in software testing -- the most common methodology for validating quality of code -- can provide a basis to develop a familiar and effective new approach for checking the correctness of designs. We lay a new foundation for testing designs in the traditional spirit of testing code. Our specific focus is the Alloy language, which is particularly useful for building models of software systems. Alloy is a first-order language based on relations and is supported by a SAT-based analysis tool, which provides a natural backend for building new analyses for Alloy. In recent work, we defined a foundation for testing Alloy models in the spirit of the widely practiced unit testing methodology popularized by the xUnit family of frameworks, such as JUnit for Java. Specifically, AUnit, our testing framework for Alloy, defines test cases, test outcomes (pass/fail), and model coverage, and forms a foundation that enables development of new testing techniques for Alloy. To provide a more robust validation environment for Alloy, we build on the AUnit foundation in four primary ways. One, we introduce test generation algorithms, which automate creation of test inputs, which is traditionally one of the most costly steps in testing. Two, we introduce synthesis of parts of Alloy models using sketching by enumeration and constraint checking, where the user writes a partial Alloy model and outlines the expected behavior using tests, and our sketching framework completes the Alloy model for the user. Three, we investigate optimizations to improve the efficacy of our core model sketching techniques targeting improved scalability. Four, we introduce a second approach for sketching Alloy models that incorporates attributes from our test generation efforts as well as our initial sketching framework and uses equivalent formulas to outline expected behavior rather than tests. To evaluate our techniques, we use a suite of well-studied Alloy models, including some from Alloy''s standard distribution, as well as models written by graduate students as part of their homework.