On May 10, 2018, an email was sent out to Course 6 students warning them that 6.813, User Interface Design and Implementation, will not be offered this spring due to the professors who are able to teach it having “conflicting teaching and sabbatical commitments.” The email also eluded alluded to the possibility that the course may not ever be offered again; instead, it will be replaced by more specialized courses.
While the idea of having more classes related to human-computer interaction (HCI) is appealing, lack of a solid plan for replacing 6.813—particularly, related classes that will satisfy the same requirements—is concerning. Getting rid of 6.813 without having suitable replacement classes in place will adversely impact Course 6 students interested in HCI, especially graduate students.
6.813 is one of Course 6’s most popular and versatile advanced classes. For the last several years, the class’s enrollment has hovered at around 200 students. For undergraduates, 6.813 counts as one of the two Advanced Undergraduate Subjects (AUS2) needed to graduate. M.Eng. students can take the masters version of the course, 6.831M, as one of their four Approved Advanced Graduate Subjects (AAGS). Finally, the doctoral version of the course, 6.831D, can count as one of the four classes needed for the Technical Qualification Evaluation (TQE) for Ph.D. students.
For undergraduates, the loss of 6.813 is disappointing but not disruptive. There are some related AUS2 classes, namely Engineering Interactive Technologies (6.810), and, potentially, the planned “new undergraduate subject on data visualization” that could be taken instead.
For M. Eng. students, losing 6.831M stings a little bit more. The only directly-related AAGS is Intelligent Multimodal User Interfaces (6.835). There are some tangentially-related courses, including Advances in Computer Vision (6.869) and Advanced Computer Graphics (6.839), that could be taken to fill the other three slots, but the fact that these count toward the same concentration as 6.831M doesn’t change the fact that they are more closely related to the topics of intelligence and performance than usability and design.
Ph. D. students interested in HCI have lost the most with the departure of 6.831D, as they are now left without a TQE class in their field. Again, there are some tangentially-related options, but HCI researchers will certainly have a harder time getting the grades needed to complete the TQE than researchers of algorithms or artificial intelligence who can take two classes directly in their area of expertise.
This is admittedly a very requirements-oriented take on the matter, but the fact is that requirements have a very strong influence on what classes students take. Instructors should be aware of what requirements their classes are actually satisfying, and administrators should take care to design the curriculum to cater to the diverse set of interests within the department.
It seems that the department heads have noticed that there are remarkably few HCI classes at MIT while there are a large number of people interested in taking them. Replacing 6.813 with several new classes would help solve this problem, so the change should be welcomed. However, the people organizing the change should be aware that 6.813 is satisfying not one but three requirements. Only once replacements are in place that satisfy all of these requirements will it be safe to discontinue 6.813. Broadening MIT’s HCI curriculum is a great idea, but we shouldn’t leave students interested in HCI with nothing to take while we do it.