Over the last few months, it’s become pretty clear to us at MakingBetter that the Experience API has finally reached some sort of tipping point. No longer do we get questions about what the Experience API is. Not only are people starting to get an idea about what it is, they’re getting a sense of what it can enable. Now, THE question we’re asked most often in emails, phone calls, first consultations, conferences, workshops — you name it — is this:
“How do I get started with the Experience API?”
Usually, this comes from a few different contexts. I’ll try and address this question and, as suggestions come in about other resources to include, I’ll curate this post.
Getting Started with xAPI… for Everyone
First, read the spec itself on GitHub. The Experience API is the largest and most successful Open Government effort in terms of the number of people truly outside of government contributing to it (like… not even contractors) and the 4th largest Open Government project overall. That’s pretty darn impressive considering the impact on other programs and agencies in US Government alone, and that it’s seen so much adoption.
The Experience API is just one technology — the first of a larger architecture we proposed to the Center for National Research Initiatives back in 2010. The information the US Department of Defense (via ADL) has available on the Experience API, and the larger Training & Learning Architecture of which xAPI is a part, broadly paints the picture of the research and development that is being performed now and into the foreseeable future, much of which will find its way to standardization.
We’ve put out a few posts and articles that should help with some general information — like, what is the value proposition of the Experience API? There’s also some narrative history of how it evolves from the last big eLearning specification, SCORM.
Widely read and very detailed, Mike Rustici’s “Layers of the Onion” is a very good read to suggest ways in which one can start using the Experience API (in your head, replace any mention of “Tin Can” or “Tin Can API” with “Experience API” and it all makes sense).
On Twitter, the following people tweet pretty knowledgeably (and regularly) about the Experience API:
- Russell Duhon (@fugu13)
- Andrew Downes (@mrdownes)
- Ben Betts (@bbetts)
- Ryan Smith (@tweetRyanSmith)
- Craig Wiggins (@oxala75)
- Jason Lewis (@canweriotnow)
Getting Started for Designers
When people ask us about getting started and they have a general sense of what we’re all talking about, there’s a difference among people who want to get the thinking around an application of Experience API before diving in — basically, they want to focus on the design of the thing first.
Among the first articles that addressed the kind of design paradigm that the Experience API enables, generally, is this one: Designing for Conscious Decisions. However, the thinking that goes into designing with the Experience API goes a bit deeper. The ties to Vygotzky’s (and Engstrom’s) Activity Theory are pretty strong. Years ago, I attempted to baseline the thinking with The Fundamental Design of Learning Activities.
Going back over a year, however, there are a couple of case studies that have undoubtedly been influential in the adoption of the Experience API we’re seeing now (and both of the following examples are in Michigan… just to point it out for reasons that will be obvious in a few weeks… more teasing). First, Ellen Meiselman is arguably the first actual adopter of the Experience API in the seminal work she’s been doing with the University of Michigan Health System. Not long after Ellen got started, Sean Putman detailed the use of Experience API to benchmark expert users of Altair Software to compare/contrast and help guide the improvement of novice users (Part 1 | Part 2).
Sean and Janet Laane Effron teamed up to put together a MOOC on Designing With Data. That MOOC is long-since over, and it’s not specifically about xAPI, but it talks to how to design with data in a conceptual manner and the materials are still available for all to go through.
While these people may not tweet specifically about xAPI (as in hardly ever), their general discourse on design — instructional, experience and/or other — are of the kind of influence we’d like to see in more and more work with the Experience API:
- Julie Dirksen (@usablelearning)
- Janet Laane Effron (@jleffron)
- Sean Putman (@SeanPutman1)
- Sarah Gilbert (@melsgilbert)
- Steve Flowers (@xpconcept)
Getting Started for Implementers
Some people really want to dive in, play with making activity statements, send them to a database or a Learning Record Store, aggregating all that data and see what they can understand around the patterns of activity that have been recorded. These people just want to get in and “start doing stuff.” I’m not a fan of this approach, but there is something to be learned by hacking and making, so let’s jump in.
If you’re not already on the mailing list(s) for xAPI, you should sign up right away. Once that’s done, here are some things you may be jumping into:
- I want to make activity statements with some web content.
But… let’s say you don’t know a whole lot about code and you don’t have a lot of content to work with. Sean Putman recently shared a set of templates at DevLearn 2014 to a standing-room-only crowd, and his templates look really easy to work with. Download the templates.
- I want to play with a Learning Record Store.
There are a lot of ways to do this. There are at least three turn-key LRS solutions offered as Software-as-a-Service and at least two solutions you can download and install behind a firewall or run locally on your own computer… for free.
All three of these solutions are pretty mature. Learning Locker is a pretty lean offering but offers good access to your data. WaxLRS has the best reporting features and custom APIs to get access to your data of any of these options. SCORMCloud doesn’t offer a lot in terms of data visualization stock, but it’s free to use as a sandbox (currently), whereas Learning Locker and WaxLRS, as SaaS, have some variable pricing depending on how much data you’re storing per month.
If you are running local on your machine alone, the ADL Sample LRS is the easiest and fastest approach to doing so. If you want to run this from a server, you’ll likely want to put the time into setting up Learning Locker. If you’re using WordPress, you may want to look into the standalone LRS recently released by Grassblade.
- I want to visualize data from the LRS
UPDATES & EDITS: