ABSTRACT. A free flowing discussion on the technology available in the University of Wyoming's SHELL 3D Viz Center. The discussion will include a brief presentation of previous projects, highlighting how students and faculty have utilized virtual reality tools such as the 3D CAVE and head mounted displays. Discussion will focus too on how students can intern in the Viz Center, how faculty can collaborate with the staff and write joint grant applications, or seek letters of support for utilizing the facility for the benefit of enhancing the broader impacts aspects of grant applications.

ABSTRACT. Stylized from the famous confidence-building story by Dr. Seuss, this BOF focuses on how can we use Virtual and Augmented Reality to improve the climate for women in computing and STEM fields. There has already been work done in using virtual humans and avatars to reduce implicit bias. This BOF will present current work in this area and lead a discussion of how this disruptive technology can be used to level the playing field. A further discussion will ensue of the advantages and disadvantages of this technology will be included. Participants will walk away from this BOF with an exchange of ideas of how we can use virtual reality, partnerships to further work, and an understanding of the pros and cons of the technology

ABSTRACT. This BOF will provide a chance for attendees to learn about applying for patents.

ABSTRACT. Research has shown the quality of the advisor-student relationship has a direct impact on the success and timely completion of the student’s research experience. Though there is not one standard plan that fits all students, finding and fostering the advisor-student relationship is key. This BoF will explore topics and strategies to enrich the journey through college and graduate school: selecting an advisor, communication strategies in meetings, working through problems, the power of mentors and peer groups, productive time with your advisor, and detecting signs it may be time to find a new advisor. Participants will walk away from this BoF with a clear understanding that it is possible – and critical - to have a good working relationship with their advisor and a plan of action to take steps towards creating one. We encourage undergraduate and graduate college students to attend; however, everyone is welcome.

ABSTRACT. Interested in participating in outreach to help support girls of all ages in computer science? This Birds of a Feather will provide an opportunity for sharing ideas as well as hearing about lessons learned from existing outreach programs.

ABSTRACT. As manifested by many of the past exposed vulnerabilities, most of those could have been avoided if better secure coding practices were in place. To help developers in security testing of their applications, static and dynamic code analysis tools have been developed. However, these stand-alone tools are normally used after whole application is developed. Often, security testings are conducted by separate security analysts, not by the developers themselves. Thus, developers remain unaware of the vulnerable codes and do not pay attention to security aspects while developing other applications. Moreover, security testing at the later stage of development adds to the cost and time of development. In the recent years, security-check plugins are being developed that can be integrated with IDEs to get in-situ security-analysis and feedback while coding. Unlike regular stand-alone security analysis tools, developers do not need to go through a separate analyzing for the whole application/code. In this talk, I will present such available IDE plugins and their capabilities. I will also review developers’ perspectives on secure coding and security tools as found by previous studies. At the end, I will discuss some future research directions that are worth exploring to understand and improve the state of security plugins and secure coding practices in general.

ABSTRACT. Many modern geo-replicated distributed systems sacrifice strong consistency for high availability. They provide weaker forms of consistency that do not necessarily require all replicas to be synchronized immediately when some data is updated. The clients connected to different replicas can submit transactions operating on the same data concurrently. Depending on the timing of updates propagation to other replicas, the transactions can be processed in different orders on different replicas. This relaxed ordering constraints on the transactions may result in subtle orders that can violate the application invariants. We present a novel dynamic exploration method for weakly consistent distributed systems that explores different schedules of transactions to detect application invariant violations. Due to the different synchronization constraints of different levels of consistency, all orderings of transactions are not allowed in a system. We adopt delay bounding to systematically generate and exercise different schedules considering the consistency guarantees enforced by the system. Our implementation and evaluation on a distributed key-value system shows the applicability of our method to detect violations.

ABSTRACT. Object Oriented Programming has been increasingly popular than conventional procedural programming paradigm with the addition of a number of additional features based on the concept of "objects"; one of such feature is information hiding. Information hiding gives objects the power to restrict access (code visibility) to its certain parts from the running code. But, it has often been discussed that information hiding might have an effect on the testability of a program. However, there has not been much research to answer how code visibility actually affects the effectiveness of automatic testing tools. This paper aims to study if there is any effect of code visibility on two major performance traits of automatic test generation tools: the code coverage of the tools and their fault finding ability. The analysis of results based on two state-of-the-art automatic testing tools on two benchmark programs showed that low code visibility does not necessarily lead to low code coverage. In addition, we used mutation testing to conclude that the fault finding ability of the testing tools is also not affected by code visibility.

ABSTRACT. In applied visualization research, artifacts are shaped by a series of small design decisions, many of which are evaluated quickly and informally via methods that often go unreported and unverified. Such design decisions are influenced not only by visualization theory, but also by the people and context of the research. While existing applied visualization models support a level of reliability throughout the design process, they fail to explicitly address the influence of the research context in shaping the resulting design artifacts. In this work we look to action design research (ADR) for insight into filling this gap. In particular, ADR offers a framework along with a set of guiding principles for navigating and capitalizing on the disruptive, subjective, human-centered nature of applied design research, while aiming to ensure reliability of the process and design. We explore the utility of ADR in increasing reliability of applied visualization design research by: describing ADR in the language and constructs developed within the visualization community; comparing ADR to existing visualization methodologies; and analyzing a recent design study retrospectively through the lens of ADR’s framework and principles. This presentation will give a high level overview of the visualization design process and will explore the potential benefits of applying an ADR approach to this research context.

ABSTRACT. Genetic Algorithms are a popular set of optimization algorithms often used to aid software testing. However, no work has been done to apply systematic software testing techniques to genetic algorithms because of the stochasticity and the lack of known outputs for genetic algorithms. Statistical metamorphic testing is a useful technique for testing programs when the output is unknown or when the program has random elements. In this paper, we identify 19 metamorphic relations for testing a genetic algorithm. We implemented 15 of these relations and show, through mutation testing, that these 15 relations are more effective at finding defects than traditional unit tests based on known outputs. We examined the failure rates of the system-level relations when initialized with various fitness functions. We examined the failure rates of the system-level relations when initialized with various fitness functions. We found three relations failed excessively and we then modified these relations. This is the first time statistical metamorphic testing has been applied for testing genetic algorithms.

ABSTRACT. Talk Abstract: This talk will be presented by Angela Benavides, a graduate student at the University of Wyoming in the 3D Interaction and Agents Research Laboratory. Benavides will present her work on an investigation of bimanual actions which inform design and implementation of 3D interaction techniques. This talk will present what we are doing to study bimanual actions and what we have learned so far to inform interaction design. This talk will serve as preparation for an upcoming research talk at conference on the following paper.

Paper Abstract: This paper presents the details of a novel selection technique, a solution to combat the challenges with volumetric data selection explored using immersive display spaces. We present results on a physical experiment that we used to inform design and discuss the details of our proposed technique. We also present the results of a virtual experiment to evaluate feasibility, utility, and usability. A description of user strategies and design is included. Our results will have implications for applications that use alternative rendering techniques as well as volume selection to assist with selection tasks in a physically-based manner. Our technique can be used in combination with other selection and manipulation techniques to improve overall desired performance of volumetric selection. Our work is relevant to any immersive interaction space which may have difficulty interacting using its current rendering form or where volumes of objects require a selection task.

ABSTRACT. Kernel density estimates are important for a broad variety of applications. Their construction has been well-studied, but existing techniques are expensive on massive datasets and/or only provide heuristic approximations without theoretical guarantees. We propose randomized and deterministic algorithms with quality guarantees which are orders of magnitude more efficient than previous algorithms. We will also describe how geometric and topological inference of a point cloud can be calculated by examining its kernel density estimate with a Gaussian kernel. This allows one to consider kernel density estimates, which are robust to spatial noise, to approximate computation in comparison to raw point sets.

ABSTRACT. Diversity continues to be a topic of great concern in Computer Science (CS). As a teaching assistant for the CS Senior Capstone Project course at my university for the past three years, I have worked with 17 women out of 104 total students (16.3%). The TA role involves meeting as mentor and advisor with the team weekly throughout the academic year, and with each student individually twice per semester. By my observations, women students are prepared academically for the rigors of the senior capstone experience. Yet I continue to observe patterns of struggle and misperception which plague them. These are some of the lessons I have learned as I strive to promote engagement of women students in the CS capstone project course.

They need to know that CS matters to me as the instructor, and what about it excites me. It isn’t enough to just focus on the foundations of CS, I need to express my feelings about computing.

They need to know that CS is more than writing code, and that there may be other facets of software engineering which truly engage them. This understanding grows as they connect with women across all phases and aspects of software engineering.

They need to feel that someone believes in them. This may be the first step to believing in themselves, or it may just be shoring up their foundation in a moment of doubt.

They need to know that Imposter Syndrome is real, and they need strategies to combat it. Especially critical to overcoming Imposter Syndrome is having the social foundation such that they can see that they are not alone in their doubts.

Culturally responsive pedagogy would have us teach as though our students’ lives matter. As we should learn to see the world through their eyes, we can better guide them toward understanding the field of CS and how it relates to them. In the process, we help ground their sense of identity and belonging with the larger community of practice, and guide their understanding of the legitimacy of their contributions to impactful project work. While I observe growth in confidence in all students as they move through their capstone experience, I believe the women students most greatly benefit from that increased confidence as they transition from school into the workplace. It is my hope that this increased confidence will strongly correlate to long-term persistence in CS.

ABSTRACT. “Women in Computing” topic has gained a lot of importance around the world with the emerging information era. Middle East is no exception. Arab women in computing (ArabWIC), is one of the most successful organizations that was founded to connect, support, inspire and promote Arab women in the technical field.

About Arab Women in Computing(ArabWIC):

ArabWIC was founded during 2012 New York University Abu Dhabi (NYUAD) regional collaborative Workshop on Women in Computing in the Arab World with NYU Abu Dhabi. Professor Sana Odeh is the founder and chair of ArabWIC. Professor Sana Odeh is an Affiliated Faculty of Computer Science, New York University Abu Dhabi, Faculty Liaison for Global Programs of Computer Science, and Clinical professor, Courant Institute of Mathematical Science, New York University, New York.

The goal of the Arab Women In Computing(ArabWIC) is to support, inspire, retain, encourage collaboration among, increase visibility of and help elevate the status of Arab women in computing/technology all sectors(students, faculty, industry and entrepreneurs), and allow them to achieve their career goals. Also, to create linkage with international women-in-CS organizations. ArabWIC is now an official chapter of the Anita Borg Institute for Women and Technology (Systers).

Within the first year, Arabwic succeeded in building a community of about 450 Arab technical women. Currently, this number has jumped to reach more than 2000 members from 17 different Arab countries. ArabWIC also held a number of successful conferences and events and has become an integral entity of Grace Hopper's Systers community.

One of the powerful tools that ArabWIC uses for achieving its goals and advancing the status of its members is the Mentoring program. Mentoring can be particularly important in communities where role models are scarce and where there may be a limited existing networks of family and friends who share their experiences. ArabWIC members face a unique combination of challenges, including an industry that is male dominated, cultural norms for women in the workplace or schools, geographically dispersed communities, and infrastructure inconsistencies among others.

In the United States female enrollment in Computer Science and Engineering has fallen from 37% in the early 80’s to a dismal rate nearing 20% today. However, as Professor Odeh’s preliminary research indicated that in the Arab world, females comprise the majority of undergraduate students studying Computer Science in Algeria, Egypt, Jordan, Morocco, Saudi Arabia, and the UAE. Professor Odeh along the collaboration with ArabWIC started to conduct a comprehensive study on Women in computing in the Arab world to explore the representation of women in IT, the opportunities, and the diverse challenges facing women in computing in the Arab world. More importantly, beta findings identify different issues women face in the field of computing in the Arab world that the US. The result will be published in a paper and a book within the next few years.

ABSTRACT. Scale-out key-value stores are simple, have low operational complexity, and handle faults well. A decade of key-value storage optimizations have given us stores that handle hundreds of millions of operations per second per machine and can scale to hundreds of machines. Research prototypes of more complex systems are now being built on top of these systems with the hope of high performance with fault-tolerance "for free".

Unfortunately, despite near-microsecond access times, the simple get/put interface of these systems suffers from two key problems that will limit the success of such approaches. First, it exposes semantics to developers that are as difficult to use as lock-free multicore programming. Second, applications are forced to make many remote requests to perform even simple higher-level operations, which harms performance.

We seek to address these issues with two key contributions. First, we will quantify the application-level performance cost of existing simple interface. Second, we are exploring a new stored procedure model for distributed key-value stores to ease application development and improve performance.

Our goal is to construct prototypes from three different domains: a transactional database engine, a real-time graph database engine, and a real-time machine learning application. These prototypes allow us to evaluate the of the existing interface and to find commonalities in the way they interact with storage. Insights gained from the measurements will be used to build a generalized stored procedure model that enables applications to push data processing to the storage system where data resides, instead of making frequent remote requests.

ABSTRACT. EvoSuite is a fully automated test case generation tool for unit testing of object-oriented software in Java. It addresses the challenge of test oracle generation. First it automatically generates a test suite using evolutionary algorithms and then suggests possible oracles by generating sets of assertions that maximize the number of seeded first-order mutants revealed by test cases. In this paper EvoSuite is studied and the effectiveness and efficiency of generated test cases on a set of nine open-source projects is evaluated and their mutation scores compared with those from MuTest. EvoSuite test cases achieve 10% higher mutation score than MuTest.

ABSTRACT. My very recently published research work involves the implementation of Proteus, a mobile network service control platform to enable safe and rapid creation and evolution of services within a mobile software defined infrastructure (one that supports ‘soft’ technologies like software-defined networking, cloud computing, network functions virtualization, etc.). With Proteus, service designers specify the details of network services in service templates which capture the network components involved in realizing the service, the relationships between such components, way these components need to be configured and managed (lifecycle functions) and service specific policies associated with the service. The Proteus orchestrator and resource placement engine then combines the generic templates with resource availability in the infrastructure and a specification of the desired characteristics of a service instance, for example, in terms of scale and location, to dynamically realize service instances.

Proteus thus ‘opens up’ current single mobile provider environment to third-party service and application providers. It allows for different instances of a service to be instantiated in parallel. Because different service instances can be executing in parallel, it also makes it possible to instantiate variants of the same service, or indeed radically different service instances. Proteus also helps with rapid and safe evolution of existing service instances, e.g. the architecture of a standards-based mobile broadband service instance can be modified to support selective offloading of delay-sensitive user traffic to an edge cloud.

To illustrate the feasibility of the the Proteus platform in dynamically deploying, managing, and allowing the evolution of network services, we realize three different services from the literature with Proteus: a broadband mobile network service based on the standards-based evolved packet core (EPC), a selective mobile offloading service [SMORE by J. Cho, et al.], and a personalized mobile-aware cloud service [MobiScud by K. Wang, et al.]. In our evaluations, we use the PhantomNet mobility testbed to serve as the software-defined infrastructure.