Last September, I reported on the Blackboard to Desire2Learn migration at SNL Online, as we moved from our initial planning and design phases to full production and review modes. I promised then I’d keep you posted; I had no idea how all-consuming the migration would prove to be.  Changing requirements, mission creep, and faculty training kept us continually challenged; here’s a look back at each:

Changing Requirements

We developed a course template through an iterative process that included a couple of rounds of user testing with faculty and students. Though we did extensive requirements gathering before developing our prototypes, because of extremely tight time constraints we did so based on our test instantiation of D2L. Platform strengths and shortcomings became known as the actual D2L production environment was installed, and we found ourselves in a long cycle of review and redesign. These redesigns came while our course migration was well underway, which resulted in having to rebuild courses already converted in a meticulous, labor-intensive process. Much of this was likely unavoidable, given we were gathering requirements and developing templates while our IT department struggled to integrate an unfamiliar and untested platform. More time to research, design, test, and iterate on a stable installation in advance of the actual migration would have prevented a lot of headaches.

Mission Creep

We designed our migration procedure to include a course-author review process; authors would review their converted course to ensure accuracy and in doing so would become oriented to D2L.  However, what had been envisioned as essentially a proofreading exercise quickly morphed into rounds of actual course revisions, which exponentially increased the workload. The courses were better for it, but once again more time to plan and allow for contingencies like this would’ve been helpful.

Faculty Training

The paradox of D2L is that the very features that are so empowering for some faculty are confusing and intimidating for others.  FITS devoted enormous amounts of manpower to create extensive documentation and video tutorials for faculty university-wide, but because of the unique nature of the SNL Online template and program we had to have specialized materials and training for our faculty. Changing user and system requirements complicated this process, and my team is still developing training materials as new needs and problems are identified.

The Rollout

Though there were problems both avoidable and unanticipated, the initial rollout of D2L in January was remarkably trouble free. Our user testing gave us a template that leveraged D2L’s strengths and that students found easy to use. Faculty unfamiliarity with the platform and its tools—despite generous training and support resources—required far more attention than student difficulties.
Overall the migration was a tremendous success.

Some problems remain. Certain D2L tools still don’t work properly in our environment; intelligent agents don’t work at all. However, the folks at D2L are dedicated and responsive, and I’m hopeful that these issues will be resolved.

We plan to do a follow-up study to determine how our design can be improved and another to test whether students do better academically with one design versus another. There’s some debate in my department on whether usability and pedagogy are sometimes incompatible; I’d love to see the results of a study addressing that concern.

In a future post I’ll look at lessons learned and what I might do differently given similar circumstances.