Der deutsche Education Blog

Microsoft Research Connections Blog

The Microsoft Research Connections blog shares stories of collaborations with computer scientists at academic and scientific institutions to advance technical innovations in computing, as well as related events, scholarships, and fellowships.

  • Microsoft Research Connections Blog

    Unlocking Academic Success with Frame Games for Learning


    Unlocking academic success with frame games for learning

    It began with a simple question: “Why can’t students earn digital rewards for being awesome?” A research group comprised of university faculty, staff, and students at the Rochester Institute of Technology (RIT) decided to find out. The team delved into the everyday travails of college life—from academia to social activities—and developed a real-world game, Just Press Play, which helps students earn a digital reward for the ultimate achievement: collegiate success.

    Each game participant receives trading cards equipped with secret codes and a radio-frequency identification (RFID) keychain that they can swipe to “check-in” at permanent and temporary locations. The cards and the keychain are just the first two tools that the students must learn to use in order to progress through the various aspects of the game. The game is based around challenges that occur both online and in the real world. The challenges are designed to encourage students to venture out of their comfort zones and get involved in all aspects of school—including interactions with school faculty and staff. Completing the challenges also gives the player access to web pages and videos that tell the story of an alternate history of RIT.

    Get Microsoft Silverlight

    Blending Technology and the Humanities

    The alternate history of RIT describes a battle fought between two rival factions: The Athenaeum and the Mechanics Institute. The Athenaeum represents the creative and exploratory aspects of a student’s academic journey and the Mechanics Institute represents technical mastery. In order to succeed, the student must understand both aspects, although they may ultimately join one side or the other.

    As the program manager who chose to fund this project, it should come as no surprise that I have a background in both science and art—and that I manage programs in games for learning as well as digital humanities. Our project team likes how RIT blends the technology and humanities disciplines. RIT focuses not only on STEM—science, technology, engineering, and math—but also on what we like to call STEAM: science, technology, engineering, arts, and math.

    Financial support for the development of Just Press Play was provided by Microsoft Research. This project is also the culmination of three years of collaboration with RIT’s School of Interactive Games and Media as part of our Games for Learning Institute. The mission of the institute is to study and create games that are fun, educational, and effective. Just Press Play fits that bill perfectly.

    Just Press Play officially launched on October 13 at the RIT School of Interactive Games and Media. The kick-off event was streamed live to project partners at the University of Wisconsin, Teachers College at Columbia, and the New York Law School. These partners are assessing the effectiveness of the program and also exploring legal issues related to gameful education.

    Next Steps

    The Just Press Play team intends to expand the game throughout RIT next year if the initial pilot is successful, but other educational institutions will have to wait a bit before trying out the technology for themselves. Just Press Play will initially be available exclusively to students enrolled at RIT. The Just Press Play team will refine the structure of the game based on early results and then roll it out to partner schools at a later date. The developers are hopeful that the lessons learned from these early “games” could eventually be expanded to include more college-level institutions and, potentially, all education starting with pre-school and extending through lifelong learning and professional training.

    The ultimate goal of the Microsoft Research Gameful Education project is to support the development of a unified game layer for education, one that can unify gameful experiences across schools and technologies. This platform will drive better educational outcomes and enable entirely new types of educational research.

    Donald Brinkman, Research Program Manager, Games for Learning, Digital Heritage, Digital Humanities, Microsoft Research Connections

    Learn More

  • Microsoft Research Connections Blog

    Night at the Museum—sans Ben Stiller


    Jonathan Fay and I recently visited the California Academy of Sciences, the world-class natural history museum and research center located in San Francisco’s Golden Gate Park. Our goals were to connect in person with Academy researchers and to help prepare the Academy’s Morrison Planetarium for an Academy NightLife event. This particular NightLife featured Microsoft technology—including Windows Mobile—and a planetarium show that demonstrated the astronomy capabilities of Microsoft Research’s WorldWide Telescope. I was particularly interested in talking with Academy researchers about my work with the new and more local phase of Worldwide Telescope, which uses recently developed tools and capabilities to create earth-science data visualizations and tell stories—supported by the high-resolution imagery of Bing maps.

    Hippocampus reidi, the Atlantic seahorse whose phylogenetic tree we are re-growing in WWT

    Hippocampus reidi, the Atlantic seahorse whose phylogenetic tree we are re-growing in WWT

    I spoke at some length with several of the Academy’s researchers about the potential of WorldWide Telescope (WWT) as an earth-science research tool, and about how to visualize interesting and often complex datasets. As one consequence, I am now working on a proof-of-concept example for them that uses WWT to render the phylogenetic tree for a particular genus of seahorse as a space-time diagram in relation to the Earth. The idea is to illustrate the genetic distance between different species in relation to spatial separations between their native habitats. By extension, such diagrams could span orders, families, and even classes of living organisms.

    We also spoke with the Academy researchers about broader data problems: specifically, the potential for using what we call the Environmental Information Framework—developed within our Earth, Energy, and Environment group—to address information-management challenges. We discussed different approaches to managing and publishing data and how Microsoft technologies (research tools, products, and services) might be applied to help.

    Another fun development from this visit: by using GeoSynth, the standalone version of Microsoft Photosynth, I generated a digital representation of the museum’s iconic Tyrannosaurus rex (T. rex) from a set of photographs I took. I then created a Worldwide Telescope narrative tour that relocated the T. rex outside to a nearby baseball park. (I have her playing center field.) The tour features several of the source photographs as well; so we are exploring a combination of different types of data to tell the story.

    To further expand on this new idea of employing WWT for more than exploring stars and galaxies: I use WWT to create visualizations of robotic submarine missions, clouds of organic molecules culled from Arctic rivers, twisted and knotted magnetic fields around the sun, distributions of soil carbon across the United States, biodiversity of sharks throughout the world’s oceans, and lots more. The complexity that all these datasets have in common makes it difficult to place them on a chart; to get them into WWT, all I need is some sort of coordinate basis—be it geographical, geometrical or parametric—and off we go!

    Digital representation of Tyrannosaurus rex with photo inset of actual skeleton
    Digital representation of Tyrannosaurus rex with photo inset of actual skeleton

    To me, the common denominator between seahorses, dinosaurs, stars, planets, and carbon molecules is our fascination with learning about the natural world and the enjoyment of sharing our understanding with one another. Scientists spend years learning the details of their specialization, learning which questions to ask next. With the emergence of new data-generating tools to answer these questions, we see the corresponding emergence of the “drowning in data” syndrome, a malady we find rampant across all specializations and domains. Well, computers got us into this fix, and so we are working on ways to use computers to get us out.

    As a member of the Microsoft Research Connections team, I feel extremely fortunate to have the opportunity to talk to research scientists and say, essentially, “How can we help?” Given a few minutes to show them what we have in the works, the ensuing conversations are enjoyable, and often lead to productive collaborations. We hope that those collaborations, in turn, lead to solutions that will be usable by others in the earth- and life-sciences. But, as a T. rex playing center field might suggest: one step at a time…so I’m looking forward to our ongoing conversations with the Academy.

    —Rob Fatland, Senior Program Manager, Microsoft Research Connections

    Learn More

  • Microsoft Research Connections Blog

    Building a .NET Quality Control Tool for Next-Generation Sequencing Technologies


    The challenge of DNA sequencing is central to all genomics research, and while the technology has existed since the 1970s, today’s massively-parallel sequencing instruments are capable of producing gigabytes of raw genomic data quickly and increasingly cheaply. Reconstruction of a DNA sequence from this data (for example, through de novo assembly) is a compute-intensive task, and experimentation has shown that data quantity is no substitute for quality when it comes to the accurate reconstruction of a DNA sequence. Unfortunately, not all sequencing technologies produce reliable and accurate results, and experimental data will always contain varying rates of error. Therefore, a preliminary quality control (QC) step is regularly employed to detect and counteract such sequencing errors.

    Sequence Quality Control Studio (SeQCoS) user interface

    The QC of sequencing results may range from simple manual filtering procedures to comprehensive automated solutions. To contribute to this area of QC tools development, we present Sequence Quality Control Studio (SeQCoS), a Microsoft .NET software suite that is designed to perform an array of QC evaluations and post-QC manipulation of sequencing data. SeQCoS generates a series of standard plots that illustrate the quality of the input data. These plots (saved in JPEG file format) provide information on commonly observed measurements, such as GC content (the proportion of guanine and cytosine nucleotide bases in a DNA sequence), and distribution of quality scores at position-specific and sequence-specific levels. In order to filter out poorly performing sequences, SeQCoS also conducts basic trimming and discarding functions to manipulate sequence files.

    At Microsoft Research, the Microsoft Biology Initiative team is collaborating with academic research groups in the sequencing of various organisms. To ensure that the sequenced sample is not contaminated by other strains or sequencing vectors, SeQCoS optionally integrates NCBI BLAST for PCs running the Windows operating system to search against a BLAST-formatted database. We provide a pre-formatted database of NCBI UniVec, a repository of vector sequences, adapters, linkers and PCR (polymerase chain reaction) primers that are used in DNA sequencing; however, researchers can use a different database if they prefer.

    About the Tools

    SeQCoS was written in C#, using the .NET Bio (formerly the Microsoft Biology Foundation [MBF]) bioinformatics toolkit and Sho, a data analysis and visualization application. It is freely available as open-source code under the Apache 2.0 license. Further details and software downloads are available from Sequence Quality Control Studio.

    .NET Bio is a library of common bioinformatics functions (file parsers, algorithms, and web service connectors) that simplify the creation of bioinformatics applications on the .NET platform and is an open-source project that is freely available for academic and commercial use under the Apache 2.0 license. While this project was initiated by Microsoft Research, it is owned by the Outercurve Foundation, a non-profit organization, and is governed by a growing community of users and contributors.

    Kevin Ha, Microsoft Research Intern

    Learn More

Page 87 of 131 (391 items) «8586878889»