We need to have a way to archive courses (in batch and individually), so that its hidden from everything.
At USU we are about to start our 15th semester on Canvas and our instance is starting to remind me of an episode of Hoarders, I'm just hoping no one is killed by a stack of falling courses. I know many of the other institutions may not see a need for it now but in the future this will become a very real problem. With the preservation of student data, these archives would be useful for among other things, grade challenges, letters of recommendation, and accreditation.
Over the past year, as the semesters courses keep stacking up, I have thought of several different scenarios that could facilitate this process.
Out of all of them this would be, I think this scenario would be the easiest route:
The basis of the archiving process would follow a similar process as deleting a course. The course would disappear from all aspects of Canvas but unlike the deletion process it would be non-destructive. In the spirit of Canvas being a Cloud-based system this is not so much of an archival process but a way to just get the courses hidden from everything. For this to work the deletion process will have to be changed to a non-destructive method. So the engineers could copy the deletion process and modify it so you can still delete a course if needed and those that need to be archived or hidden can use a similar process. Currently when a course is deleted the enrollments are removed from the course so restoration is a two step process. If the deletion method were changed to just a simple active/inactive toggle and it disappeared, we could live with having the course still reside on the system.
Control of archiving and restoring should be a task between admins and instructors so we as admins can control this on a term by term basis but the professor has the option on a course by course basis. I recall an archived idea about offline archives but there was a consensus that instructors should not have this ability. Each institution is different and we operate courses on the premise that the course is the instructors IP, so we are more stuarts of course data than having the final say in what they can and cannot have enabled. I think this would best work as a setting that admins can change themselves.
Here are some other scenarios I came up with that give the institution the ability to do offline archives:
Scenario 1: A separate custom course export is created that includes the student interactions and is downloadable. This archive would be something akin to a IMSCC package but somehow include student interaction and assignment submissions so that there is a path for restoration if the need arises.
Scenario 2: Using Amazon Glacier storage a course could be moved to this system in its entirety. I don't know enough about Glacier storage to know if the course could be replicated to a database in Glacier or if it would have to be similar to scenario 1, but the archive file is stored in Glacier. The price of Glacier storage is significantly cheaper than S3 buckets.
Scenario 3: The course export would basically be a static copy of the course as it would appear in Canvas, including student interactions and assignment submissions. This option would be a non-restorable version but it would be a complete record of the entire course. The ability to export courses as an EPUB file could play into this.