Following the ransomware attack on the Canvas learning management system that threw thousands of institutions into several days of chaos in the waning days of spring semester, we have been ensured by the ed-tech consultants of the world that “quitting” Canvas at the institutional level is rather unlikely.
Executive strategic consultant Mike Corn told Inside Higher Ed’s Kathryn Palmer, “It’s a lot of work for an institution to change their learning management system, because they have thousands of classes that need to be ported over. It’s typically a two- to three-year process.”
Phil Hill, “an education-technology market analyst,” thinks colleges would be “crazy” to develop their own LMS technology infrastructure, because if their system was hacked, they’d be more vulnerable to criticism. Apparently sticking with the herd as hackers target everyone is safer than standing alone and getting picked off.
I do not have the years of industry observation of these gentlemen, but I did put in my time teaching college, both before and after the arrival of the learning management system, and I am here to declare that it is possible to teach effectively without such technology.
Would I want to return entirely to the predigital era? No, but over the course of nearly two decades of use, I realized that some of the aspects of the LMS I thought were most desirable or necessary were actually preventing students from developing important capacities.
This is my story.
When the LMS first appeared, I was excited for an always-accessible repository for the course schedule, assignments and student grades, and then a little later, a place for students to share their work with me and each other. I was nowhere near a power user, but fairly quickly the practices I was engaged in—semester schedule in a paper syllabus, physical assignments handed out in class, completed assignments turned in to me—were replaced entirely with digitally mediated communication and exchange.
I thought the convenience and ease of access was a good thing, and we saved a lot of trees, but over time, I believe I created an unintentional culture of passivity and disengagement with the logistical requirements of the course. Maybe this seems small, but over time, it grew ever larger as a problem.
For example, in class I would display an assignment’s guidelines on the projector, read through it with students, solicit questions and remind them that they could always access it through the LMS.
After experiencing a repeated pattern over multiple semesters of students failing to follow the assignment guidelines in both big and small ways, I checked the LMS access logs and saw that fewer than half of the students had ever looked at the assignment. Was this the case back when I was handing students physical copies of the assignments? I had my doubts. Students stuffing those pages in the folder they then stuffed in their backpacks, unearthing the artifact when it was necessary, seemed to be more accessible than a document online.
Similar patterns repeated when I changed the turning in of assignments to digital-only by the start of class. When I required a physical copy in class, I would get near 100 percent completion (the exceptions being students who were having specific immediate challenges). With digital-only copies I would get maybe 50 percent submission, with another 30 percent saying that they “just forgot.” I started having to carve out five minutes at the start of class for students to follow through and submit their files.
At the same time, the number of incomplete or undone assignments rose significantly, to 10 to 20 percent of the total. On a 750-word assignment, I would sometimes get a 300-word rough start submitted, as though being able to upload something to the system was what most mattered.
Over the course of my career, readings went from me providing copies (through textbooks or course packs), to asking students to print copies from digital files, to accepting reading texts on-screen. This made the texts easier and cheaper to access, but the way we engaged with the texts changed, and not for the better. In fact, I believe fewer students read the texts ahead of time, thinking that they could get by with having it in front of them during a class discussion.
I recall early in my predigital days admiring the degree of organization some students displayed with their paper calendars/planners. Upon receiving my syllabus and schedule, they would copy every assignment and due date into the planner, creating a kind of bible of what they were going to have to do that semester. God forbid this document is lost, but the physical representation seemed to be of help to many.
In theory, the automated, fully accessible calendar of an LMS should be superior, but I’ve come to believe that this very process of information transfer of students creating their calendars was reinforcing the scope and rhythm of the work they would do over the course of their semesters.
The arrival of generative AI and its capacity to remove almost all friction from the creation of academic artifacts has given me a fresh perspective on these past experiences. I now see that my general default to using the LMS led me to remove some points of friction that may have been productive in terms of student engagement and effort.
Planning ahead to make sure you know your schedule of due dates or will have access to the reading or a printed copy of the assignment to turn in requires a degree of attention and planning not necessary in the digital space. When a student would miss a class and the only way to access the assignment was to come to my office and retrieve a physical copy, I had a chance to have a conversation with them about what was going on.
How many problems had I unknowingly headed off with 60 seconds of conversation?
When I graded physical copies, I read and responded more slowly, constrained by the limits of my horrible handwriting. Digital texts and comments significantly sped up my assessment practices, but they also proved alienating from the student texts I was reading.
When it came to the LMS, I’d fallen for the M (management) part of the equation, causing me to miss some aspects of the L (learning). Over time I started to pare back my LMS use, starting with the online grade book to both de-emphasize grades and make students responsible for monitoring their performance should they so desire.
By the end of my full-time teaching career, I stopped putting the assignments on the LMS, going back to handing them out in class or requiring a visit to my office to retrieve them. I kept the schedule and syllabus as a document in the system but did not put any of the dates into the course calendar function myself.
I will not lie: This caused some measure of turbulence with students, and had I been more dependent on positive student evaluations, I might’ve hesitated to stick with these moves, but I was fairly convinced that including these frictions required students to practice skills that were useful to them.
By considering the LMS affordances through a lens other than convenience, I had to think more deeply about why I was taking a particular action or approach.
Without intention, I stumbled on the framework of “minimal computing” recently explored at IHE by Lee Skallerup Bessette, a longtime participant in and analyst of digital learning and current assistant director of digital learning at Georgetown University. Skallerup Bessette shared the minimal computing framework of Roopika Risam and Alex Gil:
- What do we need?
- What do we have?
- What must we prioritize?
- What are we willing to give up?
It strikes me that this framework is particularly useful as a lens for considering AI use in a course as well.
I’m pleased both to now have this framework to work from going forward, and also to recognize that I could stumble into some of its insights on my own.
Maybe we can’t truly get rid of Canvas (or the LMS), but it seems like an obvious positive step if everyone up and down the institutional ladder looks at what this technology is truly good for.
