Open Education on the Move: An Interview with Vijay Kumar
Q. Open education is everywhere, from MITx  and edX , to initiatives like Khan Academy, Coursera and Udacity. How do you define open education and what's behind its tsunami-like momentum?
A. In 2001, MIT launched OpenCourseWare (OCW) , which catalyzed the movement. OCW courses are snapshots of real courses and they present good models, they can be emulated.
The following year, in 2002, UNESCO coined the term Open Education Resources (OER). OER means more than open content. It includes open applications, tools and architecture, as well as legal enablers like Creative Commons  that allow the spirit and practice of open education to be exercised.
Over the past decade, the Open Education movement has expanded dramatically. Openness has become part of the discourse about educational change, whether it's at the level of a university or a country. For example, there are over 250 institutions in the OpenCourseWare Consortium . National and international organizations like the Commonwealth of Learning  have adopted open education as a central strategy for providing quality education on a large scale.
Q. So there's amazing breadth to the movement, what about depth? How does online learning go beneath the surface?
A. First, it's important to distinguish between online learning and open education. Openness brings a lot of added dimensionality to online. Educators often refer to the Four R's — reuse, redistribute, revise, remix — as the critical attributes of openness. So it's not only about enabling unfettered access to resources but also making it possible for them to be used and adapted based on the goals of the learners.
Communities of self-learners can remix open materials and help each other learn. It's an era of connectivity combined with collectivity. People from around the world can participate, and millions have.
Q. Can you give an example of how technology fosters open learning?
A. Video lectures have become increasingly common, but how do you search them? The Spoken Lecture Browser , a technology developed by Jim Glass's group in CSAIL , addresses this issue. Built on voice recognition technology and artificial intelligence, it lets you automatically transcribe the audio portions of a video lecture and then search those videos using natural language. For example, I can do a search on "angular momentum" if I missed a few classes in introductory physics. The tool lets me find the relevant segments through a search, rather than wading through 20 hours of lectures.
OEIT is working to make this technology more widely available to our faculty, perhaps through MIT TechTV . As this intersection of technology and openness becomes broadly available, it's easy to imagine the reuse of precise segments from these lectures by another instructor or by students for self-learning. Searches on a concept could lead to related materials, either confined to the course or across the Internet, paving the way for deeper learning experiences or alternate pathways for learners with different motivations.
Q. So open education enables students to choose the mode of learning that works best for them?
A. That's the direction we're headed in. For example, alum George Zaidan, who did OpenLabWare  many years ago, recently teamed up with Professor John Essigmann in chemistry to develop online modular offerings for learning core concepts in chemistry such as "buffers" that travel across courses. This is part of the experiments in modularity  launched by the MIT Council on Educational Technology (MITCET) .
So they focus on a concept, and they might provide videos of labs to explain the concept, they might provide interactive simulations. They work with different modalities so that students have the opportunity to grok the concept in different ways.
Q. How do these online offerings impact learning in the classroom?
A. Understanding that is an important part of the MITCET experiments. A good example is the approach taken by Professors Karen Willcox and David Darmofal in AeroAstro. They said, look, if you can modularize topics traditionally taught in the classroom and offer them online, then you can have more opportunities for discussions and active learning when students come to class.
This is what's known as the flipped classroom model. A lot of the lecture and topic-related materials are posted online, and students can review these on their own time. Then, during class, the focus is on hands-on exercises.
Q. What's next for open and online education?
A. There are many exciting developments in the world of open education resources. One of note is the Open Textbook movement that is gaining legs and helping address a significant obstacle for educational access — the cost of textbooks.
Openness is being employed as a driver for addressing large problems of educational access and quality. One example of this is the Kaleidoscope Project, a project on which I'm an advisor. Funded through a Next Generation Learning Challenges  grant, it involves seven community colleges collaborating to create courses using existing OERs, with each course being developed by at least two partner institutions.
The project so far has demonstrated a substantial reduction — around 90 percent — in the cost per course per student, a one-term savings of about $60,000. The focus, however, is not only on cost-effectiveness but on improving the course design and learning results based on analysis of embedded assessments.
There's a movement to use learning analytics to improve the quality of online education. Learning analytics refers to the analysis of a wide range of data produced by and gathered on behalf of learners to assess academic progress, predict future performance and spot potential issues.
Through analytics, you reconstruct your interventions with your learners based on how they're accessing the materials, how they're performing, how they're understanding different concepts. Courses are open, but they have to be constructed in a way that you can use analytics to understand what's working and what's not. The Open Learning Initiative (OLI)  at Carnegie Mellon has a good model for this.
What's really ahead is the opportunity to learn a lot about learning — about the use of social networks for learning, about making labs widely available online, about the approaching wave of new applications.