Pin It

Why SCORM Tin Can API Will Fail (Initially)

It has been 10 years, give or take, since the last major SCORM update was introduced to the learning community.  SCORM 2004 was a great evolution from 2.1, but can the same be said about the upcoming Tin Can API?  Unfortunately, I think not.

Let me first make something crystal clear:  I want the new SCORM Tin Can API to flourish.

I think the concept behind it is a very good direction for SCORM and the learning field overall.  Will there be unforseen difficulties?  Absolutely.  But in the end, I think it will ultimately be a step in the right direction.  In fact,  it will likely become part of WordPress LMS.

Being as it may, I do think it will fail, at least at first – and by fail I am referring to the overall adoption of SCORM Tin Can by the major LMS providers, and more importantly, those who are selecting a LMS.

I have read various statements, articles, and watched videos on Tin Can, and it is clear that a revolution is in the works.  So much so that I have even seen LMS be referred to as “LRS” (Learning Record Store) – and here is where I believe the problem lies.  Change can be good, but it is difficult.  Why make it more difficult by attempting to change the structure that is in place?

The LRS reference is a minor example, but I think it proves my point.  Is there a need to relabel what we call a Learning Management System simply because we are upgrading SCORM?  Will the entire industry jump on board?  Will admins start calling themselves “LRS” administrators? Likely not.

If the advent of SCORM has proven anything, it has shown that the adoption of new standards/methodology by companies is slow process. Organizations are just starting to become educated to the benefits of SCORM, but it has taken a while.  Heck, there are countless of organizations out there that still strictly use SCORM 1.2 (which I guess is better than nothing).

I suppose what I am getting at is that SCORM Tin Can API should think twice about trying to distance itself from its predecessors.  There is comfort in familiarity, and I believe that organizations will be more ready to adopt the new SCORM protocol if it “plays nice” with its past.  If Tin Can separates itself too much from its largely establish past, then it will definately fail, at least initially.

That said, this change is a good thing (despite some of the glaring weaknesses), I personally think it will take some more trial and error before it readily becomes a norm.

 

 

Categories

About the Author:

Justin Ferriman is the Founder of LearnDash, a WordPress based LMS and Learning Strategy provider. He also works as a Learning & Collaboration Consultant where he implements large-scale training programs for Fortune 500 companies.

18 Comments
  1. Justin.

    First off, thanks for taking the time to dig into Tin Can enough to assess some things about it. We really do love when people push and prod at the work we’ve done. It gives us a sense of what’s working and what isn’t. I’m gonna jump in with some explanations of the “why”, because I think those things help sometimes.

    The choice to refer to a Learning Record Store (LRS) as something distinct from an LMS was absolutely intentional. There are two reasons for this.

    1. Tin Can is intended to define something _narrower_ than an LMS. It is one aspect of what many LMSs do, and the spec wants to acknowledge that narrowness.

    2. Learning Record Stores can exist in contexts _other_ than LMSs, and we want that possibility to be clear. In building a WordPress LMS (which is an idea I like), I’m guessing you didn’t try to tie the entirety of Moodle into WordPress, because Moodle does other stuff that wouldn’t tie nicely to WordPress. It would end up looking like Frankenstein. Instead, you want to add just the right capability to WordPress, and maybe that capability is what Tin Can refers to as an LRS.

    So, I agree, there will still be LMSs. People will still be LMS administrators. But LMSs may not be the only home for Learner Records going forward.

    You do perceive an important and open issue, though. There’s a tension between holding onto existing models and names and moving forward with technology and new names. The SCORM brand brings a real familiarity and gravitas for some. For others, the SCORM brand brings a snicker. It’s been fascinating to watch, and I can’t claim to have exactly the right answer there. Fortunately for me, I think the decision about branding Tin Can lies entirely with ADL, so I’m off the hook.

    Another great point you raise: Adoption. Specifically, getting a standard off the ground is a huge challenge; we refer to it here as the chicken and egg problem. This problem, though, is one we can really affect here at Rustici Software. Because more than 150 LMSs and 150 content providers/tools use our SCORM tools, and they’re architected to support just this kind of evolution, we will be able to seed the market with Tin Can adopters very quickly (compared to the 10 years of uptake on SCORM, at least).

    Lastly, if you want to talk weaknesses, I’m happy to do that too. (This post may be getting a little long-winded, though.) These weaknesses, or open issues, are exactly the kind of thing we’re working through weekly on the ADL webinars. In fact, I believe “verb variation” is up this Thursday.

    If you want to talk WordPress LMS at any point, I’m happy to do that too. Conceptually, our SCORM Cloud plugin for WordPress is an LRS for WordPress, nothing more. I love where you’re headed here.

    Tim

    • Justin

      Hi Tim, thanks for the comment. If I understand what you are saying, you are suggesting that the Tin Can version of SCORM will be used in conjunction with, say, SCORM 2004 3rd edition? I am not sure this is the impression I originally had of Tin Can (an “add-on”, essentially).

      So an organization could in theory make the following choice for their LMS:

      1. SCORM 2004 + Tin Can
      2. SCORM 2.1 + Tin Can
      3. SCORM 2.1/2004 Only
      4. Tin Can Only

      In the event that it is Tin Can only, organizations are receiving a narrow view of their end users learning experience, as you put it.

      If I am an organization using SCORM 2004, which is providing me item level detail, to what extent is their equal (or more?) value to include Tin Can, other than perhaps a high level reporting metric? Will organizations see the utility in this?

      In reading about Tin Can from the SCORM.com website, I can appreciate the following:

      No more back-and-forth from the content to the LMS, just send a statement to the LRS when the activity is completed.

      Seems fair enough, nothing wrong here – all good things. But to play devil’s advocate a bit: “Who cares?”

      The end result is still the same, no? The status of the event is still communicated, now it just seems to be communicated in a different way. I get that it is nice to no longer have to deal with javascript, various API calls, and so forth, but for an organization that already has a previous SCORM version implemented, what’s the benefit? How do I “sell” the idea to my CEO?

      With 1.2 to 2004, we could really see the benefit. The amount of extra detail had everyone drooling. With tin can, it’s less clunky, but less detail (at least to the uneducated)

      This kind of leads me to believe that the LRS concept is ideal for a company who is just about to implement some sort of learning system, and who wants that high level reporting. Maybe I’m a little off base here, and of course it’s all speculation.

      But as long as I am speculating, the more I look into it the more I think tin can could potentially have a drawn out adoption period. It will be interesting to see the general industry reaction.

      • I’ll pick out pieces here to try to keep it short:

        Yes, any LMS will be able to choose which standards to adopt. Where you’re missing something, though, is that Tin Can is in no way limited to the simple “Tim finished” statement. The vocabulary for Tin Can already fully expresses what is possible in SCORM. Further, the structure of Tin Can allows for _extension_, meaning additional verbs, nouns, and objects are possible. “Tim attended a conference session.” “Tim achieves a score of 72% on ‘this exam’.” Each statement can include context/detail/state. Tin Can only does _more_ than SCORM in that regard. (Sequencing, however, is being eliminated left to the content provider.)

        The two big Tin Can selling points are these:
        1. Tin Can uses a modern communication method (web services) which removes this core SCORM limitation. In SCORM, it must be a single learner, taking browser based content. In Tin Can, you can use mobile devices or simulators (a browser is not required) and you don’t have to start from the LMS. That’s a huge leap. Content can live anywhere (including real world activity).
        2. Tin Can uses a vocabulary structure that allows for widely varied statements.

        The key point is this: Tin Can acknowledges that learning happens outside of the context of an LMS session. Vast portions of learning, certainly greater than 50%, happen outside of directed learning situations. Tin Can allows for a system to capture _experiences_, and ultimately, that will allow software providers to link experience to performance.

        I agree with you on this point: Adoption will take time. I think we’ll move more quickly than many standards for a number of reasons. One of the biggest, in my mind, is the positive response from the community of vendors and buyers. People seem to realize that the experiences we’re tracking through LMSs are simply not enough.

        Getting the snowball rolling down the hill will take some time. I agree. But conditions seem to be right.

        • Justin

          Great conversation we have going here – thoroughly enjoying it.

          I am glad you cleared up some of my misconceptions re: the level of reporting, I was scared for a second there!

          Regarding the two selling points. For the first one (no longer browser bound), this makes me curious as to how industry leading content creation software like Captivate/Articulate will adapt. I suppose I don’t have any answer here, I am just trying to wrap my head around it. For engagements that I have worked on in the past, companies prefer their training to be securely hosted (https) – and therefore in a browser. Naturally, this gives them control they desire as they don’t want their new processes, tools, and so forth available to competitors. In cases like this, the lack of browser (and therefore lack of traditional control) could be seen as a liability.

          For the second point you make, it’s nice to see that there will be flexibility in the statements. That said, I find myself really getting caught up in the semantics of it all. What do we mean by “Justin Learned XYZ”? Can Tin Can (or any LMS for that matter) effectively measure learning? I suspect “learn” is used loosly here… probably more accurately described as “completed” and “passed”.

          I am reminded of my consulting gigs where I we have to present to leadership the impact of their employees’ learning. More often than not, we utilize the KirkPatrick method, which as you know involves more than measuring quiz results but includes a wide range of metrics analysis across a variety of dimensions to demonstrate learning through tangible figures. Show me a LMS that can do this and you have a 100 million dollar idea!

          • Vendors: They’re coming along quickly. A good example. Articulate will produce a native iOS app that renders their content beautifully and consistently. This is hard in HTML5. Tin Can will provide the mechanism by which they can still communicate to the LMS (which they couldn’t have done from an iOS app with SCORM).

            Protective Customers: Those who wish to continue using the browser, or https, will _definitely_ be able to do so. There are some great security options outside the browser, too, but browser will certainly be an option.

            Statement flexibility is great, yes. The concern you’re expressing is one that we’re aware of. We call it “verb variation”, and it’s going to be the topic of this week’s ADL conference call. We’ve written about it here: http://scorm.com/project-tin-can-phase-3-verb-variation/

            Tin Can is a vocabulary targeted at allowing things like you’re doing. Mostly, we want to get SCORM out of the way so that you, as instructional designers, can do what you need to. We want to let you _design_.

      • Erik L.

        Regarding “So an organization could in theory make the following choice for their LMS”
        As I understand TinCan currently, you can’t do options 1 and 2.
        You can do options 3 or 4.

        TinCan is a new API. It doesn’t seem to carry any existing SCORM spec.
        If content can be created so it reports to both, that would seem possible (complex, but possible) and would require the LMS to be able to make both APIs available…and accept calls to both APIs at the same time.

        I see TinCan as an excellent option, as it was designed, to break out of LMS-provided content. Now we’ll be able to embed learning (event!) tracking in a variety of media…like super-cookies ;-)

        I do think most ‘LRS’ implementations will be better serviced by integration with traditional LMS features to handle authentication, enrollments, transcripts, reporting, etc. However, TinCan/LRS is their simple forms will allow simple tracking for those who don’t need all those features (i.e. as is currently possible with Captivate and its simple reporting to Acrobat.com or a custom PHP script…and on that line, it shouldn’t be too hard for Captivate to extend that to TinCan as well).
        Erik

  2. I don’t think the TinCan spec will ‘fail’ but it may take a while to catch on!

    It’s a neat concept, and I understand the potential commercial success we have customers who don’t really need the whole ‘LMS Experience’ and for whom the more-simple TinCan concept may indeed work very well.

    But in its current form, it’s also much less robust. The weaknesses you and Tim both discuss are real, as well as additional items (at least in a structured LMS sense), like specific rule-based enrollments. Additionally, when will authoring tools catch up? And how robust will the authentication mechanisms need to be to allow flexibility but disallow things like hijacking someone else’s content?

    At any rate, I think you’re right there will be some lag time, which is fine. I don’t think it’ll fail, however. My biggest overall concern is if TinCan is the next iteration of SCORM itself, what’s the NEXT iteration going to be? SCORM + TinCan (i.e RTWS?) or maybe TinCan+ and an abandonment of the strengths SCORM does provide?
    Ah, change is good!

    • Fancy meeting you here, Erik. ;)

      I would actually argue that Tin Can will be more robust in a number of ways. Tin Can does nothing to discourage an LMS from controlling the experience as it does today. You can still have a user, determine what they would take, and offer it to them. Tin Can is only saying, “You can also have experience that originate outside the LMS and report back to it.” Not that you _have_ to.

      I believe that Tin Can, when using oAuth, is actually more secure than is SCORM. Few people realize it, but a competent Javascript programmer can spoof a SCORM completion. It’s possible. The oAuth workflow defined in Tin Can is definitely _more_ secure. Tin Can can know who is asserting what. That’s a big leap, too.

      Authoring tool adoption is rolling along as well. We already have seen statements coming through from companies like Rapid Intake and Articulate. At Rustici Software alone, we’ve had positive discussion with many other of the top 10 providers, several of whom are SCORM Driver customers and will have an easy adoption path. (This says nothing of the independent work that ADL is doing with activity providers.)

      I’m pretty confident that we’ve done a great job of catching the important history of SCORM in its next incarnation. We come from 8 years of working on SCORM in designing what’s to come. I’m sure, though, that we’ve missed some important stuff. That’s why discussions like these are so important to us right now. We want to prod at the spec and the work we’re doing and see what we’ve hit and what we’ve missed. So thanks to you and Justin.

  3. I I come at this from the perspective of somebody who develops scorm elearning in the rapid development tool captivate. From my point of view, tin can will provide a number of minor improvements for the traditional style elearning I develop. These include: no more frames and pop-ups which can be a headache to support; navigation and sequencing much easier to implement using html links (revolutionary!); greatly trimmed down xml files; improved control over my content and who can use it; higher level of security; option to use server side code; flexibility to report more information.

    I referred to these benefits as minor because the real selling point of tin can for me is the ability to report learning that is not traditional elearning. The statement that Bob smith passed a captivate course will be included in a feed that he was observed by his supervisor demonstrating a particular competency. Statements can be reported not just by elearning content but by the learner themselves or by supervisors, trainers, teachers, register systems, library book loan systems etc. That is where tin can will shine.

    In regards to captivate etc. supporting tin can, its not a big deal. Do a Google search for “captivate tin can” and click the top link.

    @garemoko

  4. Not a Fan

    The reason that Tim Martin is so interested in making sure that, and convincing people that, Tin Can will be adopted is that if it is adopted they will make a ton of money. They already got a grant to build it, they already had much of the architecture to support it, and if it works they will be the first and only offering available for the near term so it only stands to reason that the PR on the street from Tim and crew is that it will succeed with flying colors otherwise Rustici loses a very lucrative revenue stream and ADL gets a big fail from another one of its grants.

  5. We’re all professionals here, so I’m not sure why the notion that Tim has something to gain from the success of Tin Can somehow undermines his credibility. It’s totally cool with me if Tim earns money from the adoption of his work. I certainly hope to earn money from the work I do!

    Because my organization has long been in the business of designing training programs which incorporate offline/out of the classroom/in the field activities, I’m VERY excited about the way Tin Can opens up a better way (we do actually have a pretty good one, but it requires people to log into the learning system) to track things like manager certification of learner demonstration on the job of mastery of a concept weeks after they were exposed to the content of a lesson. It makes it that much easier to recognize and track the performance of learners as they move through the training process.

    I hope everybody wins – ADL, Rustici, content developers, learning administrators, learners, their organizations, and the customers of those organizations. This is not a zero-sum game we’re playing!

  6. Mathieu Jobin

    about the options 1-4 an organization could pick.

    Could not one LMS support both SCORM and TinCan ?
    Allowing uploads of customer’s courses files under one or the other format.
    and offering delivery in both format as well?

    it seems the way to go, so that content provider can start using TinCan without converting all of their courses. This way, the adoption should go smoothly.

    • @mathieu

      Definitely. I believe LMSs should support both SCORM and Tin Can, for an _extended_ period of time. The best LMSs will have great support for the broadest set of experiences that they can, and that will include SCORM and Tin Can.

      No, Tin Can itself isn’t backward compatible with SCORM, it does so much more. But yes, a single LMS can definitely support both elegantly.

      Tim

  7. Interesting article and comments. IMO, this API makes alot of sense- eliminating the need for Organizations, SCOS, Single-user registrations, browser requirements and connectivity. Unfortunately, the DOD isn’t too keen on updating their requirements to modern, forward-thinking standards. I’m afraid most of us will be trudging along with SCORM 2004 and LMS’s for the forseeable future.

One Ping
  1. [...] "ask an expert" option).  Interestingly, I didn't see anything that directly related to the new Tin Can API project, but I suppose an argument could be made that item #10 is kind of [...]

Leave a Reply