SCORM vs xAPI?

[helen_m_davies]

Hi all,

I'm new here, and this is my first post.  We are moving to Canvas at my institution (from Blackboard).  I do a lot with SCORM, and am looking at xAPI as well.  Do you know if Canvas has any functionality for xAPI in particular, or SCORM if not?  I can't seem to find much in the way of guides after a cursory look.

Many thanks in advance

Helen

[bneporadny]

Hi Helen,

At this time I do not believe Canvas as any functionality with xAPI. I did find this idea‌ talking about xAPI that is open for voting to gauge the interest of the canvas community‌ in wanting xapi‌ functionality. 

https://community.canvaslms.com/ideas/11980-add-a-learning-record-store-lrs-to-canvas 

[mathieu]

Thanks for the helpful responses here.  My apologies for resurrecting an old thread...

There seem to be a lot of questions and some misunderstanding around xAPI and how it fits in with LMS systems.  The linked suggestion is a good example of this.  An LMS is not meant to function as an LRS, although there are some functions in the two systems that might intersect.  This excerpt from xapi.com says it succinctly (emphasis mine):

"The LRS is the heart of any xAPI ecosystem, receiving, storing and returning xAPI statements. You’ll need an LRS in order to do anything with xAPI. Every other tool which sends or retrieves learning activity data will interact with the LRS as the central store."

So an LRS is a kind of business intelligence tool for education.  It collects, stores, and analyses learning events from a wide range of learning tools, not limited to the LMS. It is perhaps helpful to understand the working assumption of the designers of xAPI, that education happens in many different places, decentralised from the LMS.  So the LRS and xAPI are designed to federate learning from the confines of the LMS.

It seems to me that we should be encouraging Instructure to add to Canvas the ability to send xAPI event data to an LRS for analysis.  I rather surprised that Canvas does not already have any function in this regard.  Canvas DOES support receipt of results from an external tool like H5P or similar into Canvas, but this is quite limited and is in no way coordinate with a complete xAPI ecosystem.  Moodle, Blackboard, Totara, Sakai, and many other LMS's already support xapi, so it is bizarre to me that Instructure have not begun to implement what many expect to be an industry standard protocol.

I hope this helps to bring some clarity to the conversation.

[James]

helen.m.davies 

I just re-installed my own learning record store (LRS) last weekend to use with H5P and it generates a lot of data. My one activity in my stats class with 46 students generated 8.7 thousand statements for the LRS. My personal take (I've been wrong before) is that Canvas doesn't have an interest in being an LRS that would house that kind of raw data. There are other people out there who will do that, such as Scorm Cloud (not an endorsement).

The thread that  @bneporadny  linked to seems to be about getting access to Canvas' data to add to their own LRS, not getting Canvas to be an LRS. The closest that Canvas has to that is Canvas Live Events. I think that's the opposite of what you want, which is to have a learning package store information in Canvas

What Canvas will do is accept certain certain pieces of information. Currently, those are tied to submissions and it will take the grade and a submission time. The preferred way would be for someone to write an LTI that takes the information from the LRS and sends the information back to Canvas in the proper format. Scorm Cloud does that -- here's a blog post about doing it with xAPI: BLTI Dispatch using Canvas' Native External App. I have not seen anything that allows you to upload an xAPI package directly into Canvas.

Canvas does support SCORM, but the local Canvas Admin has to reach out to the Canvas Customer Success Manager to enable the SCORM LTI first. You can upload SCORM content directly into Canvas. Here's more information about it: How do I import SCORM files as an assignment?

[helen_m_davies]

Thank you James, that's helpful.