2017 Award Recipients
Dr. Pablo Calvi
School of Journalism
I teach multimedia journalism, a very intense journalism course which I created based on the Teaching Hospital model. Unlike other courses at the J-School, the students are pushed to report on the communities around Stony Brook, and on Long Island, and produce hyper-local stories which have shown to have a direct impact on our communities. The stories are all published in two magazines www.theosprey.info and www.longislander.info (they feature some of the media platforms the students learn in class, including interactive graphics, video for the web and audio, on top of text on Wordpress).
Certain sections of Long Island are media deserts. But because the stories we produce are interesting and appealing to local readers, we’ve developed and cultivated a readership that ranges between the 10,000 and the 25,000 unique visitors weekly, an unusual number for an in-class student publication. But the readers keep coming and it's become time to expand and improve our publication routines and systems and bring them to the cloud, with a better design for mobile devices and the potential to work remotely.
That’s the main reason I’m applied to the TALENT grant. In order to make this transition I'd need some actual (financial and technical) support, required to redesign these magazines, move the production structure to the cloud (mainly using Google Docs and Google Drive), and expand our range of coverage with the possibility of teaching the entire class online. As an example of our production I attach here a few stories that may give you a sense of the type of content we are producing:
Dr. David L. Ferguson
Department of Technology and Society
Colab.online: supporting collaborative learning on projects
I, like many instructors, find that, while collaborative group learning projects offer numerous hypothesized benefits, groups are difficult to manage in practice. Monitoring team interactions is costly, challenging, and often ineffective. Then, making sense of that collected data is a difficult task for instructors. Ultimately, this not only prevents us from realizing the potential benefits of teamwork, but if students are not adequately supported, we are likely teaching them to dislike and actively avoid working in teams in the future.
CoLab.online is a collaborative learning support platform that promises comprehensive support for collaborative learning group projects. A brief demonstration of the functionality showed support for self- and peer-assessment, simulated groupwork experiences to stimulate pre- and post-project discussions, gamified activities and the option to calculate ‘diversity points’ functionality to estimate and promote the value of the diversity offered by a group’s individuals. The self- and peer-assessment component is unique in that it takes a formative and non-judgmental approach. The web-based system asks students to assess themselves and their peers on a weekly basis (as opposed to at the end of the project). Furthermore, the assessment uses an innovative interface that asks students to allocate individual contributions to the group effort rather than assessing quality - a zero-sum game. Importantly, the self- and peer-assessment data is made available through a reporting interface that will offer me insight into the group’s dynamics (as perceived by the students). I would like to work with Dr. Modell, a researcher focused on collaborative group learning and author of the CoLab.online system to integrate the platform into my own courses. Some system customizing may also be necessary to meet the needs of my classes.
Dr. Paul Fodor
Department of Computer Science
online learning instruction and testing system for programming languages
The existing classroom versions of the CSE307 (Principles of Programming Languages) course often has a long waiting list of students trying to register. This proposal’s activities will alleviate the waiting list and will allow faculty to focus on improving learning outcomes. This course consist of teaching central ideas of computing and computer science, practices of computational thinking, and appropriate computing technologies as a means for solving computational problems, exploring creative endeavors, and procedural and object-oriented programming methodology. Student enrollments are over 300 students per year and we strive to provide adequate programming experience to the students by assigning problems that require a programmed solution. This practical side of the course teaches students to solve complex problems under strict requirements; document sources of code; program in an object-oriented language using concepts such as object classes, encapsulation, inheritance, and polymorphism; and use fundamental data structures such as arrays and stacks by writing sound code structure and using systematic software debugging and testing techniques.
A comprehensive online programming learning environment will be developed where students are able to receive immediate feedback on their code; identify logic errors in addition to compiler messages; access correct running solutions submitted by the instructor or other students; and view step-by-step video tutorials for extra help (in effect, making coding less intimidating and more fun). Instructor tools will include: a repository of questions for students; the ability to select problems for assignments, set deadlines, monitor and assess student learning; and an ability to determine common student errors. The assessment data collected each semester will help improve teaching and assessment practices in subsequent semesters, thereby closing the loop that connects academic assessment with effective pedagogy. Student engagement will increase through the use of a more interactive environment that can match a student’s busy schedule and provide a means through which students may repeat computer programming exercises as many times as desired. As part of this online environment, an assessment system will be designed with the purposes of improving student outcomes and meeting the requirements of the Computer Science degrees.
Dr. Thomas Graf
Department of Linguistics
using jupyter notebooks as an interactive, open source, learning platform for mathematics in linguistics
I intend to develop an interactive learning platform that teaches students the core mathematics they need for computational linguistics. The platform will greatly improve instruction at both the undergraduate and the graduate level, enhancing Stony Brook's ongoing development of Computational Linguistics curricula. By making the platform freely available online, it will furthermore help establish Stony Brook as a leader in the field of computational linguistics instruction.
Computational linguistics is a booming field in industry and academia alike, and Stony Brook has launched a professional M.A degree and has plans for a B.A. in the future. In contrast to most linguistic subfields, computational linguistics involves a significant amount of mathematics: linear algebra, abstract algebra, logic, graph theory, and formal language theory, among others. At the same time, instructional materials are scarce. Only two textbooks on the market introduce the essential math, but they still do not cover all the basics. Even more problematically, neither one is suitable for undergraduates.
n order to start offering computational linguistics courses at the undergraduate level and to strengthen our current offerings at the M.A. level, we have to create the first comprehensive yet approachable mathematics resource for linguistics students. There are four central design desiderata: 1. Application-driven: Linguistics students like language, not math, so all mathematical concepts have to be introduced through concrete linguistic problems and applications. 2. Learner-driven: The mathematical aptitude of linguistics students varies enormously. Hence the material has to be modular and flexible, giving weaker students ample opportunity to practice their skills with self-grading exercises while stronger students can dive deeper into advanced topics. 3. Interactive, multi-modal: Mathematics is hard to learn from static text, intuitions are best developed interactively from dynamic dashboards, animated figures, modifiable plots, and so on. 4. Free, open source: All materials will be freely available online, not just for consumption but also for modification. These desiderata are best met by using Jupyter notebooks, which allow static text to be combined with dynamic code snippets for user interactivity. What more, Jupyter notebooks are easily shared and can be hosted online so that they are usable on any device with a web browser.
As part of my graduate course "Mathematical Methods in Linguistics" (LIN 539) this semester I have already begun writing the static parts of the Jupyter notebooks. In the Spring semester, I will revise these static parts with a group of undergraduates as part of LIN 488. If I receive the requested funding, I will hire a graduate student to assist me in the creation of the dynamic aspects of the notebooks: dashboards, interactive plots, animations, an in particular self-grading exercises. The expanded notebooks will then be used when I teach LIN 539 again in Fall 2018 (open to advanced undergraduates), and will play a central role in developing an undergraduate version of LIN 539 to be offered in Fall 2019. Pre-post assessments and in-class surveys will be used to evaluate the usefulness of the notebooks.