Back to top

March/April 2017

Listening, Learning and Teaching, and Outreach; Teaching and Learning Computational Thinking and Algorithmic Reasoning

Krishna Rajagopal

Listening, Learning and Teaching, and Outreach

One obvious topic for this column is impacts of the new U.S. administration, including impacts on MIT, and responses at MIT: essentially every faculty member with whom I have talked over the past two months – in any context about anything – is seeing or feeling impacts, is concerned about how we should respond, and is spending considerable time reading, talking, and thinking about both. However, I do not want to write about specific developments – for example, changes to visa regulations and processes – that are changing so fast that they may morph between when I write this column and when you read it. I am also not going to write about our shared, and enduring, values. They serve us well; indeed, we rely upon them and we embody them in many things that we do and say. But, since I have no sense that we are questioning our values, and since this column is a conversation among ourselves, writing about them does not seem needed.

Instead, in the first half of this column, I have decided to share some examples of things that I have seen faculty members doing that have been prompted by the current national moment but that are at the same time, in my book anyway, things that we really should be doing in any circumstances. Please email me with other examples.

What are faculty doing?

First, listening. Many faculty have been talking, in gatherings of all sizes from two people on up. One of the most common things I have heard people talking about boils down to the importance of listening in addition to talking. Listening to people with varying perspectives and experience. Listening, with attention, to our differences. Listening to students. To staff. To faculty colleagues. Listening is always important, but it has struck me how many have stressed its importance recently. In my experience most of the time most of us are doing this well, but each of us can nevertheless look for further opportunities to hear and share differing views, and to listen.

Next, learning and teaching. These are of course central to what we as professors do, so it is natural that when faculty members respond to national events we ask ourselves how we can better understand their origins, context and consequences, and how we can provide our students with the tools to do the same.

— It is clear that there are major long-term social, political, and economic issues in America that require close attention. In response, members of the MIT community have launched Mens et Manus America, a non-partisan initiative that is convening a series of research-informed lectures and discussions to explore these issues, providing each of us with opportunities to learn. Our SHASS and Sloan colleagues leading this initiative aim to frame key questions including: What can MIT do to help address current challenges in the U.S., and bolster the health of our democracy?

— As I write this column, the most recent Mens et Manus America event was a talk by the sociologist Arlie Russell Hochschild about why some Americans vote in ways that may, to others, seem against their self-interest. (The video and a report are available here.) As you will see, Hochschild has listened in a way that few of us can and, consequently, she has things to say about U.S. political divisions and actions for finding common ground from which many of us can learn. We can all look forward to learning from future events in this series.

— What about teaching? We have a shared responsibility to prepare our students as citizens, sending them out into the world with the tools they need to play a role in strengthening civil society and to recognize when it is at risk of corrosion. Only some among us can rise to the challenge of teaching to the present national moment. But, all of us are advisors and mentors, formal and informal, whether for undergraduate students, graduate students, or postdocs. As we advise undergraduates on class selection, colleagues in SHASS have provided us with a new Web page, consisting of a curated listing of SHASS classes related to current social, political, and economic issues in the U.S. I have heard more than one colleague ask how we can best advise our students when they ask for classes along these lines; this Web page will help us, and our advisees. SHASS has also developed another listing of resources that should be of interest to all of us and to those whom we mentor at any level, called 21st Century Citizenship: Resources for Understanding and Engagement.

— Although it may already have passed by the time that you are reading this, let me mention that April 18 is MIT’s second annual "Together in Service" day and that, in addition, a group of students, faculty, and staff have planned a "Day of Engagement/Day of Action" devoted to engaging with the political, economic, and social challenges facing us.

And, outreach. We have at this point seen only a sketch of the first budget from the new U.S. administration. However, many aspects of the priorities signaled in this sketch (and in other ways) with regard to research and innovation, science and technology, the humanities, social sciences and the arts, are of deep concern. Here, we know that these are foundations upon which America builds everything from economic growth to national security, from cures for diseases to new industries and infrastructure, as well as new ways to strengthen our society and sustain our environment. Most of us are still in the early stages of planning how we can respond, but much that I am hearing falls under the rubric of redoubling our efforts at outreach, via many means, to better get the word out that building these foundations, and investing in the next generation who will develop them further for the generation after, has been and continues to be so important for the nation, and the world. This is outreach that we have long seen as important but that we may not have prioritized in our overly full day-by-day, semester-by-semester, lives. Some among us can reach out via diverse media. Others are planning visits to the offices of our representatives in Washington or are helping our students to do so. Many faculty are thinking about outreach to friends and neighbors, to business leaders whom we know, to schools, and within civic organizations and the communities in which we live. Remembering that each of our students and postdocs connects to their own home community, we should  support them in whatever outreach they may be doing. As I hear from you, I may share examples of initiatives, even at the one or few faculty scale, in my next column. Let’s see what we can do, together.

* * * * * * * * * *

Teaching and Learning Computational Thinking and Algorithmic Reasoning

Last April, Dean for Undergraduate Education Denny Freeman and I charged a working group consisting of Profs. Eric Grimson (EECS; Chair), Deepto Chakrabarty (Physics), Michael Cuthbert (Music and Theater Arts), Peko Hosoi (Mechanical Engineering), Caitlin Mueller (Architecture), James Orlin (Sloan), and Troy Van Voorhis (Chemistry) with conducting an in-depth study of what algorithmic reasoning and computational thinking mean in the context of the education of MIT’s undergraduates across all five Schools. After having incorporated substantive feedback from many students and faculty in response to an early draft, the Working Group has now completed its final report.

I encourage all faculty members to read the report in full. It is the product of a sustained deliberation. The Working Group gathered, considered, debated, and synthesized multiple perspectives. Their resulting analyses and conclusions are different from, and stronger than, anything that any individual could have done.

As you will see from their report, the Working Group explored how faculty and students across the full breadth represented by our five Schools use computational thinking; the intellectual frameworks employed in computational thinking and algorithmic reasoning; and the extent to which these are already being taught. They also found wide agreement across MIT for a set of topics that would be valuable for students to understand.

As one conclusion of their study, the Working Group recommends that MIT should acknowledge algorithmic and computational thinking as an explicit expectation of all our graduates, as they believe that it should play a role for students in all parts of the Institute.

The Working Group describes several important reasons why every MIT undergraduate should be articulate in computation as a mode of thought and a means of communication. I expect that different readers will resonate most strongly with different components of the Working Group’s argument. In my case, I value the distinctions they draw between mathematical and computational thinking. And, I follow their observation that because computers are transformational agents in the twenty-first century, our students should be cognizant of the impact of computation on their fields and should graduate from MIT as technological citizens of the twenty-first century with an understanding of the paradigms of computing – just as we expect them to be cognizant of the paradigms of (for instance) biology.

That said, what struck me most in their argument was the way that the Working Group sees computational thinking as a mode of communication. Developing a successful written or oral presentation crystallizes initially vague ideas into a tight logical argument or crisp description. And, one of the best ways to crystallize an initially vague idea for how to solve a problem is via the discipline of formulating and coding an algorithm that a computer could execute. Articulation of one’s ideas in a manner such that a computer can execute them requires precision, clarity, and logical rigor. So does communicating ideas comprehensibly to other people. The converse makes the analogy particularly sharp: the two best ways that I know by which I can shatter the belief that I understand how something works (when that is in fact not the case) are to try to teach it to a good student who questions everything or to try to formulate it as an algorithm that can be coded.

As you turn this analogy over in your mind, you start to realize that it has possible implications for how computational thinking could be embedded throughout our curriculum in ways that are analogous to how we think about teaching communication via CI-H and CI-M subjects in tandem. Just as we expect our students to learn to write well in their CI-H subjects (and not just to be proficient with grammar) we would like our students to learn foundational concepts in computational thinking and algorithmic reasoning well (and not just how to code). This comes through loud and clear in the Working Group report. The report also explicitly lays out the advantages of providing our students discipline-specific experiences with computation, the computational analogue of the motivations behind our CI-M’s.

The Working Group has considered how to accomplish the goals they articulate for the computational education of MIT undergraduates, and has recommended two options as worthy of further development. Connecting computational thinking to domain-specific contexts across different intellectual disciplines is essential. Therefore, in both of the options favored by the Working Group, at least some elements of computational thinking would be taught in subjects that are designed for a major, or designated as suitable for a major. In these ways, students would see computation in the context of a discipline that appeals to them, thus increasing the utility of what they learn. In one of the options, this would be preceded by a requirement to take one of a small group of six-unit introductions to computational thinking offered at different levels for students with different backgrounds.

Combining a common foundation and discipline-specific instruction would be comparable to the combination of CI-H and CI-M classes in the communication requirement, while the ability to embed six units of computational thinking within a larger class also finds an analogy in some of the ways in which our students satisfy the Institute Laboratory Requirement.

I hope that the recommendations of the Working Group spur faculty and departments to develop subjects that use and teach computational thinking in the context of their major, along the lines described in the report. Funds to support development of such courses will be available from Dean Freeman’s office as early as this summer. Information about applying for funding can be found here.

I am grateful to the Working Group for the sketches they have developed that provide options for how a requirement in computational thinking might be implemented, and for laying out the advantages and challenges of each approach. Further development is now required. For example, the Working Group notes that options for incorporating a computational requirement should avoid adding a significant burden on our students and recommends that a careful major-by-major study be done of the impact of allocating a REST subject to computation when considering that as a potential implementation path. They also note that, for engineering majors, the impacts on ABET accreditation of implementation options for adding a computational thinking requirement should be analyzed. The Committee on the Undergraduate Program (CUP) is the appropriate Standing Committee of the Faculty to consider these questions and, more generally, to consider how best to proceed. The CUP has now begun this work.

Although much remains to be done, I am most grateful to the Working Group for bringing this important Institute-wide discussion to this point. Their analysis, findings, and recommendations provide all of us with the impetus to take important next steps. Here again, let’s see what we can do, together.

MIT Faculty Newsletter, Vol. XXIX No. 4, March/April 2017