About the Talk
March 23, 2018 3:30 PM
Chicago, ILChicago, IL
"Delivery of educational content is evolving out of analog textbook publishing into a space where content providers must deliver a wide array of digital products, facilitating end-to-end instructional services and enabling better learning. At McGraw-Hill Education, this change involves the development of a comprehensive learning science platform that can manage the classroom experience. The classroom experience spans the range of initial rostering for a class to advanced analytics providing insights into student learning processes and outcomes for instructors. The platform needs to provide and manage all the information required to author, discover, and deliver content. Additionally, it must produce assessments, manage classroom administration, adapt lessons to the individual learner’s needs, and measure all aspects of the learning experience.
These requirements are implemented across multiple educational applications, developed by multiple teams with go-to-market deadlines dictated by the academic calendar. Successfully building this kind of platform presents the challenge of, “How do you manage information across such a large complex environment, while maintaining an information model with semantic integrity?” Teachers, parents, students, administrators and academic designers will all benefit from delivery of critical analytics. The arising challenge is to perform modeling that enables a dozen software teams to work as flexibly and effectively as possible.
In this presentation, we will discuss how they adapted more traditional waterfall based information architecture approaches to a more agile engineering environment while still maintaining strong semantic control over the information model which enables powerful analytics. This approach incorporated:
• Creating information contracts between teams • Using low fidelity models and sketches • Creating frameworks that facilitate more autonomy in individual engineering teams • Developing a baseline of customer-centric information needs to use as guiding principals • Adopting a just-in-time iterative design approach • Working problems from the outside-in, instead of from the ground up • Using collaborative design/review processes such as “hack-a-thons” and collaborative, transparent documentation"