There is a lot of hype around the learning APIs, and I wanted to write a practical explanation that I honestly wish I had before we invested the time and resources we did. Hopefully, this will help you make an informed decision.
First, let’s start out by simplifying. Basically Tin Can API, Experience API, xAPI are evolutionary names of the same API. It has struggled to get a foothold in the market place for good reason, so the name has changed over the years to re brand it.
SCORM was first on the scene and still runs the majority of e-learning courses today. When first coming into elearning industry I was immediately turned off to see SCORM 1.2 was released in October 2001 and the standard did not seem to change with wide adoption for year. Relating the API to a programming language, I immediately started looking for something newer thinking it had to be better. 10 years in the web development industry is considered to be per-historic.
So we gravitated to newly released Tin Can API and comparing the APIs on the surface immediately thought the features were so much better with TinCan. The major stand outs were TinCan could work across devices and TinCan gathered much more detail.
But as they say, experience comes after you needed it.
So we started the project by building the necessary LRS (Learning Record Store) and were immediately shocked by the recommended hardware for most of the software out there. It really did not register till we saw the system in action, so we went with a dedicated server and started the long list of necessary packages.
We went with the guys at Learning Locker who were very helpful at getting everything setup. The most daunting part of the task was setting up the necessary nonSQL database which we went with mongoDB for our setup. Start to finish, it took about 3 weeks to get running and the bugs worked out.
To the content makers, TinCan was nothing more than a checkbox selection when they went to publish their course. Articulate Storyline, required TinCan to work with their mobile app. So it was nothing but a plus for them from the course development side.
From the coding perspective, there was the added complexity of connecting the content to the new LRS, but once the library worked for one course, we could easily share that code across the others.
But it wasn’t till the student started coming on board did we get the full context of what was happening in this API.
To use an analogy, let’s compare the APIs to a home security system. For the SCORM-run house, there is just sensor on the door and a motion sensor in the room you walk into. On the TinCan-run house, there are not only the door and motion sensors, but there are multiple cameras in every room constantly recording.
If you were watching the data being sent from those 2 security systems, you can imagine you would not need much storage and bandwidth when the SCORM house only having to record the door opening and closing and any trips to the motion sensor. Meanwhile,for the TinCan system, you will be filling hard drives with months of surveillance video multiplied by all the number of class participants.
And when we watched the bandwidth and servers fill, we asked, “What’s the value here?” Well the content creators said, well we have mobile accessibility. Well not so fast there. The app that Storyline used was just a way of getting around the Adobe Flash problem on iOS. So technically, you had ability to see content on a mobile device, but it was extremely slow and that typically just resulted in support calls. We would just direct the students back to their desktops anyway.
The bloat of checking the LRS constantly started to slow down all the classes. In our case, we had about 200 simultaneous users in the e-learning class during the peaks. We immediately had to increase the server running the LRS just to make it through training peaks.
So hosting and systems sellers love TinCan. You definitely need a dedicated server with a large disk to run it. I was shaking my head wondering why some charged $1,000 a month to just run an LRS, but after setting one up and running, that became reasonable.
But practically speaking, what was the real value TinCan offered over running SCORM? Of course TinCan collected way more data points, however, when I asked the content creators, so what data point are you using now that you were not using in SCORM, we heard crickets.
To use that security system analogy, who was going to go through hours and hours of video footage looking for some kind of pattern. That is the daunting task of analyzing the TinCan’s “Big Data”.
Now, if you are large institution who has run programs for years and are looking for that level of detail then TinCan offers something that SCORM never will. But practically speaking, 99% of the online program just need to record when someone enters the class and if they passed the course at the end. For that SCORM 1.2 is all you really need.
Love to hear your comments and any stories you may want to share. Please comment below with your questions or comments.