2011 DevLearn – Beyond SCORM – Aaron Silvers

Editor's Notes

• #2: This isn’t SCORM 101This isn’t how to implement SCORMThis is about the future of SCORMShould we change it?
• #3: Why are you here?What do you want to get out of this presentation?Who do you train and how do you train them?============Who is an instructional designer / game designer?Who is a developer?What else are you?Who knows what SCORM is?Who has implemented SCORM?Who has had problems implementing SCORM?
• #4: SCORM has been great for online trainingThe friction it has removed has enabled literally thousands of companies to work togetherWorld wide adoptionDe Facto industry standard, that only happens because it works well for the problem it solves
• #5: -There are many things it does well, but namely the web-based page turner-It does what it was intended to do….the things that were commonplace ten years ago
• #6: E-learning today: page turnersOften not much more engaging than plain textNot necessarily a standards problem, but they don’t help either
• #9: ---web browser dependency---javascript---temporal session---single user---no concept of instructor / facilitator---constant internet connection presence---data model must be shoe-horned---attempt-centric nature doesn't lend itself to playing until you improve---assets must be delivered to and installed on LMS---any tracking implies results submission---general incompatibilities----tied to a single learning event
• #10: ----who the players are----they are communicating and working together (mostly)----each solving a different piece of the puzzle----we feel your pain, solutions are addressing the roadblocks
• #11: Create a learning experience APIThree phasesWrapped up formal outreach, presenting preliminary conclusions and design for feedback now
• #12: Keep It SimpleGive Me PowerHow can we make everybody happy?
• #13: Simplicity and Power are inherently at odds with one another, how to overcome?Core Design Tenants---simplest case,  get somebody up and running 8-8, 8 pages and less than 8 hours---enable complexity for those who want it, but hide it from those who don't need it---put burden of complexity on those most able to handle it and those who need it---remove constraints of traditional SCORM---backwards compatibility where it makes sense
• #14: Simple English sentence captures the essence of a learning experienceEnglish is a great example of simple by powerful and extensible
• #15: The essence of all this mess it at it’s core I Did This
• #16: API structured as a simple expression of Noun Verb ObjectProposed TCAPI is a technical binding to this simple conceptConceptually, each of these words is an extensible bucket of different concepts to expressWill come back to the tech stuff if you want, but first let’s look at what this structure can express
• #17: Hey, I'm reporting my own dataUseful at times, although in this example, we might not want to take Mike’s word for it
• #18: Wait, there's more than on person in Crew #8
• #19: watch one, do one, teach one – learning is about doing different things
• #20: this means Mike actually did what he was trained to do, ties performance data back to learning
• #21: Hark, a simulation
• #22: Learning happens outside the computer
• #23: Two things to notice that are implicit in those statements that represent very different assumptions from today’s model:On the Fly Creation – LRS can receive records for people and experiences it didn’t previously know about“This” is not tied to the actual content, but rather a pointer to the content. The thing reporting the result doesn’t have to be the actual thing delivering the experience.Think about how this relates to simulations and scenarios
• #25: Starting to form a transcript of my recorded learning experiences
• #27: Other things to noticed that result from the technical architecture we are considering
• #28: What we’re ironing out now.Complexity #1: sentences aren’t always quite this simple. Did you noticed I cheated on some of them?How do we express all of the bottom boxes without introducing unnecessary complexity?How do we express sub-sentences and related details?Complexity #2, creating a technical binding, defining verbs, providing extension points
• #30: Using statements like “I did this” and borrowing from what we see happening with Facebook and Friendfeed before it seems like a pretty renegade thing to do, but this has very strong ties to a lot of other things other working groups are doing, and it all goes back to what the Soviets did with a little thing called “Activity Theory.”
• #31: Activity Theory was developed as a way of understanding and shaping a workforce, which was of course a very soviet thing to do. Lev Vygotsky was a psychologist in the Soviet Union in the 1920s and 1930s. Games people cite Vygotsky for a little thing he came up with called the “Zone of Proximal Development” but along with that, he developed an activity theory that was heavily centered on tool mediation and the relationship of a single actor to an object or objective. It goes way deeper than this, but this is how it got started. While Vygotsky’s work in Thought and Language was pretty well known outside of the USSR, it wasn’t until the 1980s when his activity theory was known to the outside world.
• #32: That’s when a guy by the named of Yrjö Engeström proposed an alternative model to the Soviet Activity Theory that contained three interacting entities – the individual, the object and the community instead of just the individual and the object.Engeström also borrowed some ideas from Human-Computer Interaction theory. For instance like the notion of rules. All in all, this expanded Activity Theory by putting the actor in the context of a community where there were social norms and roles at work. This basic framework could help to explain social development, organization, culture, and social systems at various scales and degrees of inspection.
• #34: Trying to take advantage of a strong association with concepts from Activity Streams
• #39: Tin Can is an ADL BAA that is purely R&D in all ways. Goal was to take a blank slate and propose a solution for tracking learning content with today’s technologies on today’s internet. Tin Can phase I results will be prototyped by internal ADL folks moving forward to look at integration points between Tin Can and other Next Gen SCORM activities (I’m actually scheduling this now). We just don’t know exactly how it fits yet, and that’s ok. It’s a proposed solution based on an R&D project. We are about to prototype it with our own stuff and see how it fits. Perhaps the next step will be a formal best practice OR submission to a spec body for formalization. However, we need further R&D before we can do any of that. CMI 5 is a natural evolution of the legacy CMI 4 and SCORM specifications.  It is a formal spec/std effort by AICC that ADL is contributing to (by serving as a catalyst with our prototypes).  It addresses 75% of the most common issues found in our outreach (tin can user voice, help desk, one on one meetings, etc).  CMI 5 will provide a piece of the new communication framework in next gen SCORM. CMI 5 uses the concepts from a previous project, the LETSI RTWS with one small, but important, change: no specified binding… so how you communicate (REST,JSON RPC, SOAP,etc) is fine.  We also defined some of the missing parts of LETSI RTWS by hooking it to CMI 5. CMI 5 took the lessons learned from RTWS and uses the modified IEEE schemas… so… I see it as a more refined RTWS not an alternative to RTWS.One of the lessons we learned from SCORM is that committing all our resources to solutions without them being driven by the "market" resulted in a lot of angst. This is something we (ADL) are working hard to avoid going forward.We think agility is good. That is the goal: fast fast fast.  I foresee many different services and technologies that will eventually make up a whole “store” of next-gen SCORM services that work together in a chaordic system, but don’t have to be all adopted in lock-step, which means implementers would be able to design learning technology solutions that meet the unique needs of their stakeholders, and the data can find it’s way, when it needs to, from one service to another.  And this brings us back to where ADL is with TinCan, and how it particularly fits with Next Generation SCORM.Now that we know you *can* use the TinCan API for learning content, we’ll work on *how* that actually works in the real world… how it integrates with other technologies. Tin Can = radically different… and just not done yet.  CMI 5 – uses similar concepts to old SCORM, with updated bindings, so its ready for prime time a bit earlier.One question that certainly needs to be explored is what are the “verbs” that describe how people learn? Learning Registry is tackling this, too. Eventually we may well have controlled vocabularies but everything is new.And TinCan, in that way, is completely awesome, because we got an approach from the ground-up that meets the needs today that isn't hamstrung by legacy thinking.
• #40: Watch for adlnet.gov which will have more information on Next Generation SCORM, videos and prototypes