An LMS for Traditional Revolutionaries

One of the lessons that I learned from my days helping faculty improve their teaching with technology is that making teaching and learning effective is hard work. That's no startling revelation I'm sure, but it is an important reminder that if we want to see education change for the better, it's not going to be easy.


Discovering a better method of teaching and learning doesn't automatically lead to change across an institution; you develop it slowly, one teacher at a time. I found that there is a golden path to change: accept teachers’ individual methods and encourage them to incorporate just one or two new teaching techniques (or tools) into their current practice. Then you can help them expand, iterate, and even experiment beyond known "best practices."

When teachers begin to be successful with their new methods you encourage them to share their own unique successes (and failures) with their colleagues. This is how you develop a community of instructional innovation.

I've learned a similar lesson working for Canvas over the past three years: making a learning platform that is effective for everyone is a Herculean task. Building any learning platform that is both progressive and capable, efficient and experimental requires deliberate consideration of priorities—and hard decision-making. If you prioritize your development solely upon the requirements of institutional RFPs and user requests, you'll simply recreate the LMS of the past. If you ignore institutional requirements and user requests, and focus on the more innovative, even radical approaches, you risk alienating the majority of teachers who are confident in their current practice. If you don't accept teachers where they are at today, you'll never develop that community of innovation tomorrow. But is it possible to build technology that both respects the past and welcomes the future?

Michael Feldstein describes a similar sense of frustration in his recent post, Dammit, the LMS, addressing LMS vendors, LMS users, and LMS critics all in one go. It's a spot-on commentary, and reflects many of the frustrations I myself have had both inside and outside academia. As a member of university staff, my frustrations with the LMS stemmed from the functional limitations and outdated technology of the status quo. As a team member at Instructure, I am sometimes frustrated by people who want to see Canvas fit only tightly constrained models of learning. Canvas is different, on purpose.

Sure, we do a lot of things that conventional LMSs do, either because those things simply work for teaching and learning (retrieval practice, reflective activities, collaborative spaces, meaningful feedback, etc.), or because those things are demanded by our community.

We also do a lot of things that no one thought an LMS could or should do, and indeed they don't always immediately pay off in usage. For example, Michael references a conversation with Brian Whitmer about the relatively low percentage of teachers who use some of the open capabilities that are in Canvas. Here's what our data show:

 Canvas Capability
 Percent  Count 
 "Public" courses  3.8%  18,397
 CC licensed courses  0.6%  3,011
 Assignments that allow URL submission  4.0%  247,059
Table 1. Open capabilities used in "active" Canvas higher ed courses in the past 12 months.
Courses N = 488,493; Assignments N = 6,125,015

Now, those percentages are super-low and I think they generally reflect the level of demand from teachers and institutions for more open capabilities in education technology. But let's put this into perspective.

First, we see higher rates of adoption of Canvas than previous LMSs, meaning more faculty are using the platform, including those who have not used online tools in the past. This suggests that these results may be watered down while newcomers figure out what works best for them and what the system has to offer.

Second, when looking at use of the URL submissions option (which suggests teachers are encouraging students to create outside the LMS) keep in mind that 4 percent of Assignments over just a 12-month period is nearly a quarter of a million assignments​. And 4 percent of active courses is more than 18,000. So while the percentage is a bit sad, the actual volume is significant. Indeed, the volume is encouraging, especially to those of us who believe there is real value in open educational practices (e.g., encouraging and empowering learners to create and share from their own spaces, outside the LMS, be it a personal website, wiki, or blog).


These are just a few examples of capabilities in Canvas that we believe add flexibility and encourage different approaches to teaching and learning. I recognize that sharing this data is a little risky; some may use it to argue that Canvas shouldn't worry so much about the small percentage of educators who may take advantage of these fringe capabilities. After all, won't teachers who are actually invested in open educational practices just eschew the LMS for their own platforms anyway?

Focusing only on "users like us" and ignoring the others may work in the short-term, but for long-term success you have to build bridges, not walls.

To help education improve itself for all teachers and learners we have to try to connect with those teachers who aren't comfortable with radical shifts in pedagogy or technology. We believe that the best way to encourage positive change in educational practices across the broad landscape of content areas, learning objectives, and teaching philosophies is by providing tools that are easy-to-use, flexible, and comfortable to the majority of teachers and learners. The door to change must be open and the doorkeeper must be deposed.

Some of the ways we do this is by having an open community, engaging with people who disagree with us, and investing in the open platform aspect of Canvas. We need both traditionalists, critical pedagogues, progressive researchers, and open educators to contribute to Canvas.That doesn't have to be done through pull requests or by building LTI apps or integrations, though that's a brilliant way to build solutions that are right for your context. But by dialoging what works in teaching and learning and what doesn't. By debating what technology is best for, and when it leads us away from our shared goals of teaching and learning better in an open and connected world.

Keep Learning,
Jared Stein
VP Research & Education